EVALUATION OF LIVE PHYTOPLANKTON AND LIVE ALGAE SUBSTITUTE NURSERY CULTURE. Jefferey W. Babuin. A thesis submitted to the. School of Graduate Studies

Transcription

1

2

3

4

5 EVALUATION OF LIVE PHYTOPLANKTON AND LIVE ALGAE SUBSTITUTE DIETS FOR FEEDING JUVENILE OYSTERS (CRASSOSTREA GIGAS) IN NURSERY CULTURE by Jefferey W. Babuin A thesis submitted to the School of Graduate Studies in partial fulfillment of the requirements for the degree of Master of Science (Aquaculture) School of Fisheries, Marine Institute Memorial University of Newfoundland July, 2009 St. John's Newfoundland and Labrador

6 Abstract The aim of this study was to mitigate summer mortality of juvenile Pacific oysters (Crassostrea gigas) by developing a low-cost, low-intensity system of supplemental feeding of juvenile oysters that could be used during times oflow natural phytoplankton production. Three experiments tested the effectiveness of supplemental feeding of juvenile oysters in nursery culture using live algae and live algae substitute diets. Experiment 1 tested five dietary formulations consisting of two Schizochytrium-based live algae substitute diets (Algamac 3050 and Rotimac ) and /sochrysis galbana (clone T-Iso) in 1:1 (dw:dw) and 100% formulations. Oysters were fed for 60 days and the Algamac 3050 /T-Iso diet produced the highest percentage increase in dry weight over 60 days (+ 106%) followed by the reference T-Iso diet (+65%). Experiment 2 tested the effectiveness of using protease absence/presence as an indication of dietary suitability. The oysters that were fed the diets that produced poor growth (Algamac 3050, Rotimac and the starved control) also failed to produce proteases. Experiment 3 tested three dietary formulations consisting of two mixed Schizochytrium-based live algae substitute (Algamac 2050 and Rotimac ) and live algae diets (using phytoplankton grown in outdoor bloom tanks dominated by Thalassiosira nordenskioeldii; 1:1 ( dw:dw)) and a reference 100% T nordenskioeldii diet. Both mixed diets produced a significant increase in mean oyster dry weight and overall energy storage levels compared to the T nordenskioeldii diet following 30 days of feeding. The results of the three experiments suggest that juvenile oyster energy storage levels and overall health could be improved through a supplemental feeding regime.

7 ~ ~ Acknowledgements This project was funded through AquaNet, a Network Centres of Excellence in Aquaculture grant (339857). In-kind support was provided by Pearl Sea Products, Aquafauna Biomarine and Unique Seafarms. I would like to thank my committee members: Dr. J.N.C. (Ian) Whyte, Dr. G.J. (Jay) Parsons, Cyr Couturier and Dr. Neil Ross for their assistance, guidance and most of all patience. I would also like to thank Norma Ginther, Laurie Keddy, Kathleen Sherry, Nicole Brun for their assistance and advice. Lastly, I would like to thank my wife, Tatiana, for her never ending support. ll

8 Table of Contents Abstract Acknowledgements 11 List of Tables v List of Figures v1 1.0 Introduction I 1.1 Oyster aquaculture in British Columbia Summer mortality of juvenile oysters and probable causes Summer mortality of juvenile oysters at Sykes Island farm site Previously studied effects of toxic algae on Crassostrea gigas larvae and adults Schizochytrium-based diets Evaluation of alternative tests for determination of diet efficacy Schizochytrium-based diets and bloom tank algae Objectives Materials and Methods Experiment I Experimental setup II Summary of diets tested Algal culture Biochemical analysis of oysters and test diets Statistical analysis Experiment Experimental setup Analysis of protease activity Analysis of zymography gels Experiment Experimental setup Summary of diets tested Algal culture Biochemical analysis of oysters and test diets Statistical analysis Results Experiment I Efficacy of diet uptake over a 6-hour feeding cycle Experimental temperature Shell height, dry weight and ash-free weight Crude protein content Crude lipid content Crude mono/oligosaccharide content Crude polysaccharide content Experiment Protease activity Experiment Experimental temperature Survival Dry weight and ash-free dry weight Crude protein content 37 Ill

9 3.3.5 Crude mono/oligosaccharide content Crude polysaccharide content Crude lipid content Fatty acid content 4.0 Discussion 5.0 Conclusions and Recommendations 6.0 References 7.0 Tables 8.0 Figures IV

10 List of Tables Table 1. Proximate composition of diets used for feeding Crassostrea gigas juveniles for 60 days at 1% dry weight dai' Mixed diets were fed at 1:1 (dw:dw) and total oxidation was assumed for calculating total energy. Algamac 3050 and Rotimac are Schizochytrium-based dry feeds (Aquafauna Biomarine, Hawthorne, CA, USA). T-Iso was the flagellate species Isochrysis galbana (clone T-Iso). Table 2. Proximate composition of diets used for feeding Crassostrea gigas juveniles for 30 days at 1% dry weight dai' Mixed diets were fed at 1:1 (dw:dw) and total oxidation was assumed for calculating total energy. Algamac 2050 and Rotimac are Schizochytrium-based dry feeds (Aquafauna Biomarine, Hawthorne, CA, USA). Thalassiosira nordenskioeldii was the dominant species (>90%) in bloom tank algae used for feeding. Table 3. The fatty acid composition of test diets (n = 3; ± SE) fed to juvenile Pacific oysters (Crassostrea gigas). The dominant phytoplankton species (>90%) in outdoor grown bloom tank algae cultures was Thalassiosira nordenskioeldii. Rotimac and Algamac 2050 are spray-dried Schizochytrium-based diets (Aquafauna Biomarine, Hawthorne, California, USA). Table 4. Individual Pacific oyster (Crassostrea gigas) dry weight after 60 days of feeding test diets. Diets were batch fed for six hours at 1% of the initial dry weight of oysters per tank per day. The reference diet used was Isochrysis galbana (clone T-Iso) and Algamac 3050 and Rotimac are spray-dried Schizochytrium-based diets (Aquafauna Biomarine, Hawthorne, California, USA). Table 5. Individual Pacific oyster (Crassostrea gigas) AFDW after 60 days of feeding test diets. Diets were batch fed for six hours at 1% of the initial dry weight of oysters per tank per day. The reference diet used was Isochrysis galbana (clone T-Iso) and Algamac 3050 and Rotimac are spray-dried Schizochytrium-based diets (Aquafauna Biomarine, Hawthorne, California, USA). Table 6. The fatty acid composition of juvenile Pacific oysters (Crassostrea gigas) fed three test diets (n = 3; ± SE) at the start of feeding (Day 0) and following 30 days of feeding. The dominant phytoplankton species (>90%) in outdoor grown bloom tank algae cultures was Thalassiosira nordenskioeldii (B loom). Rotimac and Algamac 2050 (AM 2050TM) are spray-dried Schizochytrium-based diets (Aqua fauna Biomarine, Hawthorne, California, USA). v

11 List of Figures Figure l. Photograph of the tank setup for Experiment l. Each plumbed circuit consisted of a top tank with three upwellers holding juvenile oysters (Crassostrea gigas) above a reservoir tank using a mechanical pump to recirculate seawater during feeding. When feeding was not taking place, ambient 1 J.lm-filtered seawater was allowed to flow through the top tank. Figure 2. Relative feeding rate of juvenile Pacific oysters (Crassostrea gigas) as measured by the percentage reduction in cell counts of five diets. Diets were fed at 1% the initial dry weight (mg) of oysters per tank for a period of 6 hours. Algamac 3050 and Rotimac are both Schizochytrium-based live algae substitute diets and the reference diet used was Jsochrysis galbana (clone T-Iso). Figure 3. Mean dry weight (mg ± SE) of individual Pacific oysters (Crassostrea gigas) following 60 days of feeding test diets. Oysters were fed 1% (dw) of the initial dry weight of oysters per tank. The reference diet used was lsochrysis galbana (Clone T Iso) and Algamac 3050 and Rotimac are spray-dried Schizochytrium-based diets (Aquafauna Biomarine, Hawthorne, California, USA). Figure 4. Percentage(± SE) protein content of juvenile oysters (Crassostrea gigas) fed rnicroalgal and live algae substitute diets. Juvenile oysters were batch fed 1% of the initial dry weight of oysters per tank for 60 days. The live algae reference diet was lsochrysis galbana (clone T-Iso) and Algamac 3050 and Rotimac are spray-dried Schizochytrium-based live algae substitute diets (Aquafauna Biomarine, Hawthorne, California, USA). Figure 5. Protein content expressed as mg per individual dry oyster(± SE) following 60 days of feeding test diets. Diets were fed at 1% of the initial total oyster dry weight per tank and the reference diet used was lsochrysis galbana (clone T-Iso). Algamac 3050 and Rotimac are spray-dried Schizochytrium-based diets (Aquafauna Biomarine, Hawthorne, California, USA). Figure 6. Percentage(± SE) lipid content of juvenile oysters (Crassostrea gigas) fed rnicroalgal and live algae substitute diets. Juvenile oysters were batch fed I% of the initial dry weight of oysters per tank for 60 days. The live algae reference diet was lsochrysis galbana (clone T-Iso) and Algamac 3050 and Rotimac are spray-dried Schizochytrium-based live algae substitute diets (Aquafauna Biomarine, Hawthorne, California, USA). Figure 7. Lipid content expressed as mg per individual dry oyster(± SE) following 60 days of feeding test diets. Juvenile oysters were batch fed 1% of the initial dry weight of oysters per tank for 60 days. The live algae reference diet was Isochrysis ga!bana (clone T-Iso) and Algamac 3050 and Rotimac are spray-dried Schizochytrium-based live algae substitute diets (Aquafauna Biomarine, Hawthorne, California, USA). VI

12 Figure 8. Percent(± SE) mono/oligosaccharide content of juvenile oysters (Crassostrea gigas) fed microalgal and live algae substitute diets. Juvenile oysters were batch fed l % of the initial dry weight of oysters per tank for 60 days. The live algae reference diet was lsochrysis galbana (clone T-Iso) and Algamac 3050 and Rotimac are spray-dried Schizochytrium-based live algae substitute diets (Aquafauna Biomarine, Hawthome, California, USA). Figure 9. Mono/oligosaccharide content expressed as mg per individual dry oyster(± SE) following 60 days of feeding test diets. Juvenile oysters were batch fed l % of the initial dry weight of oysters per tank for 60 days. The live algae reference diet was lsochrysis ga/bana (clone T-Iso) and Algamac 3050 and Rotimac are spray-dried Schizochytrium-based live algae substitute diets (Aquafauna Biomarine, Hawthorne, California, USA). Figure l 0. Percentage(± SE) polysaccharide content of juvenile oyster ( Cra sostrea gigas) fed microalgal and live algae substitute diets. Juvenile oysters were batch fed l % of the initial dry weight of oysters per tank for 60 days. The live algae reference diet was lsochrysi galbana (clone T-Iso) and Algamac 3050 and Rotimac are spray-dried Schizochytrium-based live algae substitute diets (Aquafauna Biomarine, Hawthorne, California, USA). Figure 11. Polysaccharide content expressed as mg per individual dry oyster(± SE) following 60 days offeeding test diets. Juvenile oysters were batch fed 1% of the initial dry weight of oysters per tank for 60 days. The live algae reference diet was Isoch1ysis galbana (clone T-Iso) and Algamac 3050 and Rotimac are spray-dried Schizochytrium-based live algae substitute diets (Aquafauna Biomarine, Hawthorne California, USA). Figure 12. Polyacrylamide gel zymography of homogenised tissue from juvenile oysters ( Crasso trea gigas) at Day 0 of a 60 day feeding experiment. Each lane represents tissue from three oysters (n = 3). Low molecular weight marker proteins were used to estimate protease size (kda). Figure 13. Polyacrylamide gel zymography ofhomogenised tissue from juvenile oysters ( Crassostrea gigas) at Day 20 of a 60 day feeding experiment. Low molecular weight marker proteins were used to e timate protease size (kda). Day 20 oysters. Each lane represents tissue from three oysters (n = 3). Algamac 3050 and Rotimac are Schizochytrium-based spray-dried live algae substitute diets ( quafauna Biomarine, Hawthorne, California, USA). lsoch1ysis galbana (clone T-Iso) was used as a control diet of known quality for feeding juvenile oysters (C. gigas). Juvenile oysters were batch fed 1% of the initial dry weight of oysters per tank for 60 days and mixed diets were presented as 1:1 (dw:dw) formulations. Vll

13 Figure 14. Polyacrylamide gel zymography ofhomogenised tissue from juvenile oysters ( Crassostrea gigas) at Day 20 of a 60 day feeding experiment. Low molecular weight marker proteins were used to estimate protease size (kda). Each lane represents tissue from three oysters (n = 3). Rotimac is a Schizochytrium-based spray-dried live algae substitute diet (Aquafauna Biomarine, Hawthorne, California, USA). lsochrysis galbana (clone T-Iso) was used as a reference microalgal diet of known quality for feeding juvenile oysters (C. gigas). Juvenile oysters were batch fed I% of the initial dry weight of oysters per tank for 60 days and mixed diets were presented as 1 : 1 (dw:dw) formulations. Control oysters were starved to bacterial level (l J.. un filtration of incoming seawater). Figure 15. Polyacrylamide gel zymography ofhomogenised tissue from juvenile oysters (Crassostrea gigas) at Day 60 of a 60 day feeding experiment. Low molecular weight marker proteins were used to estimate protease size (kda). Each lane represents tissue from three oysters (n = 3). Algamac 3050 and Rotimac are Schizochytrium-based spray-dried live algae substitute diets (Aquafauna Biomarine, Hawthorne, California, USA). lsochrysis galbana (clone T-Iso) was used as a control diet of known quality for feeding juvenile oysters (C. gigas). Juvenile oysters were batch fed 1% of the initial dry weight of oysters per tank for 60 days and mixed diets were presented as 1:1 (dw:dw) formulations. Figure 16. Polyacrylamide gel zymography ofhomogenised tissue from juvenile oysters ( Crassostrea gigas) at Day 60 of a 60 day feeding experiment. Low molecular weight marker proteins were used to estimate protease size (kda). Each lane represents tissue from three oysters (n = 3). Algamac 3050 and Rotimac are Schizochytrium-based spray-dried live algae substitute diets (Aquafauna Biomarine, Hawthorne, California, USA). lsochrysis galbana (clone T-Iso) was used as a reference microalgal diet of known quality for feeding juvenile oysters (C. gigas). Juvenile oysters were batch fed 1% of the initial dry weight of oysters per tank for 60 days and mixed diets were presented as 1: l (dw:dw) formulations. Control oysters were starved to bacterial level (1 J.lm filtration of incoming seawater). Figure 17. Dry weight (mg ± SE) per individual Pacific oyster (Crassostrea gigas; n = 30) batch fed once per 24 hours test diets at 1% of the initial dry weight of oysters per tank for 30 days. Thalassiosira nordenskioeldii was the phytoplankton species dominant (>90%) in outdoor grown bloom tank cultures. Rotimac and Algamac 2050 are spray-dried Schizochytrium-based spray-dried live algae substitute diets (Aquafauna Biomarine, Hawthorne, California, USA). Figure 18. Percentage organic content or ash-free dry weight(± SE) of juvenile Pacific oysters (Crassostrea gigas) batch fed test diets at 1% of the initial dry weight of oysters per tank for 30 days. Thalassiosira nordenskioeldii was the dominant phytoplankton species (>90%) in outdoor grown bloom tank cultures. Rotimac and Algamac 2050 are spray-dried Schizochytrium-based live algae substitute diets (Aquafauna Biomarine, Hawthorne, California, USA). Vlll

14 Figure 19. Percentage(± SE) crude protein content per individual dry Pacific oyster ( Crassostrea gigas) batch fed test diets once per 24 hours 1% of the initial dry weight of oysters per tank for 30 days. Tha/assiosira nordenskioe/dii was the phytoplankton species dominant (>90%) in outdoor grown bloom tank cultures. Rotimac and Algamac 2050 are spray-dried Schizochytrium-based live algae substitute diets (Aquafauna Biomarine, Hawthorne, California, USA). Figure 20. Crude protein content (mg ± SE) per individual dry Pacific oyster (Crassostrea gigas) batch fed test diets once per 24 hours I% of the initial dry weight of oysters per tank for 30 days. Tha!assiosira nordenshoeldii was the phytoplankton species dominant (>90%) in outdoor grown bloom tank cultures. Rotimac and Algamac 2050 are spray-dried Schizochytrium-based live algae substitute diets (Aquafauna Biomarine, Hawthorne, California, USA). Figure 21. Percentage (± SE) crude mono/oligosaccharide content per individual dry Pacific oyster ( Crassostrea gigas) batch fed test diets once per 24 hours 1% of the initial dry weight of oysters per tank for 30 days. Thalassiosira nordenskioeldii was the phytoplankton species dominant (>90%) in outdoor grown bloom tank cultures. Rotimac and Algamac 2050 are spray-dried Schizochytrium-based live algae substitute diets (Aquafauna Biomarine, Hawthorne, California, USA). Figure 22. Crude mono/oligosaccharide content (mg ± SE) per individual Pacific oyster ( Crassostrea gigas) batch fed test diets once per 24 hours 1% of the initial dry weight of oysters per tank for 30 days. Thalassiosira nordenskioeldii was the phytoplankton species dominant (>90%) in outdoor grown bloom tank cultures. Rotimac and Algamac 2050 are spray-dried Schizochytrium-based live algae substitute diets (Aquafauna Biomarine, Hawthorne, California, USA). Figure 23. Percentage(± SE) crude polysaccharide content per individual Pacific oyster ( Crassostrea gigas) batch fed test diets once per 24 hours 1% of the initial dry weight of oysters per tank for 30 days. Thalassiosira nordenskioeldii was the phytoplankton species dominant (>90%) in outdoor grown bloom tank cultures. Rotimac and Algamac 2050 are spray-dried Schizochytrium-based live algae substitute diets (Aquafauna Biomarine, Hawthorne, California, USA). Figure 24. Crude polysaccharide content (mg ± SE) per individual Pacific oyster (Crassostrea gigas) batch fed test diets once per 24 hours 1% of the initial dry weight of oysters per tank for 30 days. Thalassiosira nordenskioeldii was the phytoplankton species dominant (>90%) in outdoor grown bloom tank cultures. Rotimac and Algamac 2050 are spray-dried Schizochytrium-based live algae substitute diets (Aquafauna Biomarine, Hawthorne, California, USA). Figure 25. Percentage(± SE) crude lipid content per individual Pacific oyster (Crassostrea gigas) batch fed test diets once per 24 hours 1% of the initial dry weight of oysters per tank for 30 days. Tha!assiosira nordenskioeldii was the phytoplankton species dominant (>90%) in outdoor grown bloom tank cultures. Rotimac and Algamac IX

15 2050TM are spray-dried Schizochytrium-based live algae substitute diets (Aquafauna Biomarine, Hawthorne, California, USA). Figure 26. Crude lipid content (mg ± SE) per individual Pacific oyster (Crassostrea gigas) batch fed test diets once per 24 hours l% of the initial dry weight of oysters per tank for 30 days. Thalassiosira nordenskioeldii was the phytoplankton species dominant (>90%) in outdoor grown bloom tank cultures. Rotimac and Algamac 2050 are spraydried Schizochytrium-based live algae substitute diets (Aquafauna Biomarine, Hawthorne, California, USA). Figure 27. Energy equivalent ofthe proximate composition of the organic fraction of juvenile Crassostrea gigas fed test diets for 30 days. Total components is the sum of tested components (protein, lipid, polysaccharide and mono/oligosaccharides; Joules mi 1 "oyster" 1 ). Oysters were batch fed once per 24 hours 1% of the initial dry weight of oysters per tank for 30 days. Tha/assiosira nordenskioeldii was the phytoplankton species dominant (>90%) in outdoor grown bloom tank cultures. Rotimac and Algamac 2050 are spray-dried Schizochytrium-based live algae substitute diets (Aquafauna Biomarine, Hawthorne, California, USA). X

16 1.0 Introduction 1.1 Oyster aquaculture in British Columbia Pacific oyster aquaculture in British Columbia currently consists of the purchase or rearing of seed oysters (- 1-2 mm) from hatchery facilities for ongrowing in ocean-based systems using floating upweller systems (FLUPSY), lantern nets, suspended trays, cages or by seeding beach tenures. In all cases, oyster growth is constrained by the natural production of phytoplankton. Phytoplankton abundance fluctuates throughout the annual production cycle as environmental conditions are continually changing. British Columbia's coastline is situated in the Northern Temperate climate zone and seasonality of phytoplankton production occurs due to shifting environmental conditions such as day length, temperature, rainfall, and nutrient availability (Harrison et al. 1983). A critical period in the annual production cycle occurs between the February - May diatom bloom and the September - October diatom bloom (Harrison et af. 1983; Haigh et al. 1992) when vulnerable juvenile Pacific oysters must survive on energy stored during the February - May diatom bloom event. Although the winter period of low natural food availability is longer (late October - early February), the lower water temperature during the winter period causes the dormant Pacific oysters to enter a hibernation-like state that results in a lower rate of energy utilisation (J.N.C. Whyte, pers. comm.). Therefore, the elevated temperature and lack of food make the mid-summer (May - September) period a potential source of mortality for vulnerable seed oysters with compromised stored energy levels.

17 The usefulness of supplementing the natural diet of the juvenile Pacific oysters during this critical time would be demonstrated in maintenance of optimum energy levels between bloom events and by promoting growth and increased survival. The advantage to the grower would be seen in increased production and efficiency through decreased seed purchases and a larger percentage of original seed growing to a marketable size. Brown eta/. (1998) noted similar problems and potential advantages of supplemental feeding in the temperate climate zone of Tasmania. Brown and McCausland ( 1999) found that supplemental feeding of juvenile oysters in hatchery culture had the potential to increase the cost-effectiveness of rearing because it is less expensive to feed oysters prepared diets when compared to mass reared phytoplankton systems. Costs for maintaining live algae culture systems range from US $160 to US $400 kg- 1 and algal production is estimated to account for approximately 30% of a hatchery's operating costs (Coutteau and Sorgeloos 1992). 1.2 Summer mortality of juvenile oysters and probable causes Summer mortality of juvenile oysters in nursery culture is a challenge faced by British Columbia Pacific oyster (Crassostrea gigas) aquaculturists. This problem is not a new phenomenon, nor is it a problem that is only experienced in the Northeast Pacific. Summer mortalities have been documented in the Puget Sound region of the Northeast Pacific, France, New Zealand, and Japan (Cheney eta/. 2000). Several hypotheses have been proposed as to the cause of the mortality events, including high levels oftoxic phytoplankton species resulting in low dissolved oxygen levels (Cheney eta!. 2000), 2

18 elevated summer water temperatures resulting in resorption of the gonad (Perdue 1983), accelerated gonadal development also due to elevated water temperatures and high nutrient levels (Imai et al. 1965; Tamate et al. 1965), and a bacterial disease, nocardiosis, was also proposed as a possible cause (Elston et al. 1987; Friedman 1991 a; Friedman 1991b). The two hypotheses regarding resorption of the gonad and rapid gonadal development both highlight a fundamental problem, a lack of sufficient nutrition. If insufficient food is available, the oyster will begin to resorb the glycogen and lipid stores in the gonad in an effort to fuel catabolic and anabolic processes. Also, rapid gonadal development requires sufficient energy from food sources to first fuel metabolic and anabolic processes, then to allow excess energy to fuel gonadal development. Both of these situations require that excess energy be available and, without sufficient energy, both processes were theorised to have caused the observed mortalities (lmai et al. 1965; Tamate et al. 1965; Perdue 1983). 1.3 Summer mortality of juvenile oysters at Sykes Island farm site The aim of my study was to gain a better understanding of the causes of summer mortalities of juvenile Pacific oysters (Crassostrea gigas) experienced by the original industrial partner Pearl Sea Products. Pearl Sea Products (now Mac's Oysters) is a largescale Pacific oyster (Crassostrea gigas) aquaculture company based in Sechelt, British 3

19 Columbia, Canada with its main farm site located in the vicinity of Sykes Island in Jervis Inlet, also in British Columbia ( ' N, ' W). The project was designed based on observations by Pearl Sea Products' staff over successive growing seasons during which large mortalities of seed oysters <6 mm in size seemingly coincided with the mid-summer drop in naturally occurring phytoplankton (J. Manders, pers. comm.) common to the temperate marine regions of the Pacific Ocean (Harrison et al. 1983). Two hypotheses for this phenomenon were: a) An insufficient food supply of wild phytoplankton for juvenile oysters <6 mm in shell length (based on anecdotal information and farm records indicating juvenile C. gigas larger than 6 mm experienced markedly higher survival) resulted in critically low energy reserves. b) The occurrence of potentially harmful phytoplankton species possibly coinciding with low food abundance resulted in the observed mortalities. The mortalities were thought to occur as a result of insufficient natural feed in the water to sustain and promote growth and to allow the juvenile C. gigas to accumulate sufficient energy stores to survive the extended winter period of low phytoplankton abundance. Therefore, the original goal for the first experiment was to test the effectiveness of supplemental feeding of juvenile oysters susceptible to summer mortality with local 4

20 phytoplankton species grown in bloom tanks and commercially available live algae substitute diets as partial and whole substitutions for live micro-algae. Previously studied mortality events were concerned with adult oysters; however, Pearl Sea Products' farm records indicated that the mortalities were isolated to juvenile oysters mainly <6 mm in size (J. Manders, pers. comm.). Scientists from the Shellfish Pathology Unit at the Canada Federal Department of Fisheries and Oceans (DFO) Pacific Biological Station facility examined moribund oysters from Pearl Sea Products' Sykes Island site and no primary pathogen was detected. Hinge ligament disease, Cytophaga-Iike bacteria (CLB), was present in some samples; however, it was not present during all of the mortality events and only affects weakened oysters (G. Meyers, pers. comm.). 1.4 Previously studied effects of toxic algae on Crassostrea gigas larvae and adults Crassostrea gigas larvae were shown to be susceptible to high levels (1 0 7 cell L- 1 ) of some toxic phytoplankton species, Alexandrium tamarense, A. taylorii, Gymnodinium mikimotoi, and Heterocapsa circularisquama in a past study (Matsuyama et al. 2001). The same study demonstrated that at lower cell densities (:S l 0 6 cells L- 1 ) only A. taylori and H. circularisquama caused mortality in C. gigas larvae. Alexandrium minutum was fed to adult C. gigas and it was demonstrated that there is an initial period of approximately 8 to 15 days where oyster feeding activity was reduced. However, following introduction of a diatom species, markedly higher feeding activity resumed (Lassus et al. 1999). The reduced feeding effect caused by toxic phytoplankton exposure 5

21 coupled with elevated temperatures could cause a rapid loss in energy reserves as the oysters utilise stored energy in feeding but lack quality food species to acquire additional energy. The oyster mortalities during the winter period were observed to be lower and this could be attributed to the oysters' lower metabolic activity and lack of food species allowing the organism to enter a state of aestivation. A parallel research project focussing on phytoplankton community structure and dynamics was directed towards evaluating the effects of two potentially toxic species as the cause of the mortalities (D. Cassis, pers. comm.). It was hypothesised that the summer mortality experienced at Pearl Sea Products' Sykes Island site was not caused by disease, low dissolved oxygen levels, resorption of the gonad, or accelerated gonadal development. Mortality caused by the presence of toxic phytoplankton species also seems unlikely as the species present are larger than juvenile Pacific oysters can ingest (J0rgensen 1996; Ropert and Goulletquer 2000). It was hypothesised that the mortality events were caused by the lack of nutritious phytoplankton species for a prolonged period, coupled with elevated water temperatures leading to the rapid dissipation of energy reserves. This weakened state could have caused death by starvation alone or by an unknown stressor or factor not sufficiently pathogenic to kill healthy oysters. In Tasmania, Brown and McCausland ( 1999) found that food availability was the main constraint to growth at a commercial C. gigas hatchery utilising natural algal production by pumping unfiltered seawater through larval and juvenile rearing tanks. 6

22 1.5 Schizochytrium-based diets Schizochytrium-based diets were chosen because they are available commercially at a much lower cost than cultured algae (-US $28 per kg), they are easy to use and store, and have sufficient levels of nutritional components in comparison to commercially utilised live algae species (Coutteau and Sorgeloos 1992). Schizochytrium diets upon resuspension in seawater yield particles of an appropriate size (3-1 0 ~-tm) for oyster consumption (Somin et al. 1988; J0rgensen 1996) although a previous study by Baldwin and Newell (1995) suggested that Crassostrea virginica larvae ingestion rates in nature are regulated by food quantity and nutritional quality and are less dependent on particle SIZe. Previous studies have examined the Algamac 2000 Schizochytrium-based diet as a partial supplement for live algae in hatchery-reared juvenile Crassostrea gigas (Boeing 1999; Brown and McCausland 2000). Boeing (1999) demonstrated a significantly higher growth rate using a 40% Schizochytrium substitution versus a l 00% Tetraselmis suecica diet. Brown and McCausland (2000) tested Algamac 2000 as a supplement to a natural phytoplankton diet in a flow-through nursery culture system and found it to be approximately 50% effective when compared with the Isochrysis galbana (clone T-Iso) control diet. The mussel species, Mytilus galloprovincialis, was shown to be able to subsist on a 100% Schizochytrium diet (Onal and Langdon 1999; Whyte 2003). 7

23 In Experiment 1, the effectiveness of the live algae substitute diets were evaluated as whole substitutes and partial substitutes and further experiments were designed. The goal of the additional experiments was to test a method of supplementing juvenile C. gigas in a commercial nursery culture system in an effective and efficient manner. 1.6 Evaluation of alternative tests for determination of diet efficacy Experiment 2 was intended to aid in future feeding studies by developing alternative methods of assessing the effectiveness of a test diet. Therefore, it was not intended to be directly relevant to the overall project objectives. Experiment 2 was originally designed to test the use of biochemical markers (e.g., HSPs, enzymes) as indicators of dietary stress in juvenile C. gigas. It was hypothesised that oysters that were presented with an ineffective diet would display signs of stress such as production of HSP70 stress proteins or an altered profile of digestive enzymes. It was further hypothesised that if the tests showed that an ineffective diet could be demonstrated through the production of a stress response, future feeding studies could possibly be shortened as lengthy growth and biochemical measurements determinations would be unnecessary. Experiment 2 focussed on assessing protease activity profiles. A growing oyster that is receiving an effective diet will stimulate protease activity for tissue production. An ineffective diet may cause the oyster to lower or alter protease production as protein assimilation and growth ceases. Protease activity was correlated with growth to determine if this qualitative method is appropriate for the assessment of diet 8

24 ~ --- effectiveness. Other methods of assessing metabolic activity were used in past studies with other species (e.g., Mytilus californicus) such as calorimetry, oxygen consumption, RNA:DNA ratios and the activity of key metabolic enzymes (e.g., citrate synthase and malate dehydrogenase) (Wright and Hetzel 1985; Dahlhoff et al. 2002); however, protease activity was not measured in these studies. 1.7 Schizochytrium-based diets and bloom tank algae The objective of Experiment 3 was to develop and test methods and diet formulations for supplementing a natural phytoplankton diet that could be used in remote locations similar to that of Pearl Sea Products' Sykes Island FLUPSY (Floating Upweller System) operation. Experiment 1 (E 1) demonstrated the effectiveness of a 1:1 ( dw:dw; dry weight/dry weight) substitution of Algamac 3050 and Rotimac with a natural phytoplankton diet (T-Iso). Algamac 3050 was detennined to have unnecessarily high lipid levels and Algamac 2050 was tested instead. As original plans specifying the use of live phytoplankton cultured in bloom tanks on site and field trials were not possible at the appropriate time (June- September 2002), the experiment was moved to Unique Seafanns' hatchery location where operating outdoor bloom tanks existed. Therefore, live phytoplankton cultured in outdoor bloom tanks was used as feed for nutritionally compromised juvenile C. gigas supplemented with Algamac 2050 and Rotimac. 9

25 Using the results from Experiment 3, a determination was made on the effectiveness of the theoretical bloom tank live algae and substitute diet formulations for application in remote oyster culture locations. 1.8 Objectives The main objective of this research project was to design and test a system of supplementary feeding of juvenile C. gigas in nursery culture that could be used in remote locations. The goal of the development of such a system would be to enhance juvenile oyster storage energy levels prior to the summer growing season, limiting summer mortality of nutritionally-depressed oysters. This was achieved through the completion of three objectives: 1. Determination ofthe effectiveness oftwo live algae substitute diets as whole and partial replacements for a live algae diet. 2. Evaluation of qualitative protease activity analysis as a tool for assessing dietary effectiveness. 3. Testing of bloom tank algae and two live algae substitute diets as mixed diets for juvenile C. gigas. 10

26 2.0 Materials and Methods 2.1 Experiment Experimental setup This experiment tested two Schizochytrium-based dietary formulations: Algamac 3050 and Rotimac that were previously untested for feeding Crassostrea gigas. Unlike the previous studies by Boeing (1999) and Brown and McCausland (2000), this experiment used 50% substitutions for the live algae lsochrysis galbana (clone T-Iso) and assessed growth and biochemical parameters in determining the diet's effectiveness. Isochrysis galbana (clone T-Tso) was used as a reference diet due to its excellent nutritional profile, ease of culture and wide use by industry and in previous research studies (Helm and Laing 1987; Whyte 1987; Coutteau and Sorgeloos 1992; Brown eta/. 1998). Table I summarises the nutritional information for each of the test diets. The pure Schizochytrium sp. diet, Algamac 3050, is currently used for enriching Artemia and rotifers in larval finfish culture as it is very rich in lipid, specifically, the long chain polyunsaturated fatty acids docosahexaenoic acid (DHA; 22:6ro3) and eicosapentaenoic acid (EPA; 20:5ro3) {Aquafauna Biomarine 2005; Table I). Both fatty acids were implicated in a previous study to be of key importance for C. gigas (Langdon and Waldock 1981 ). Rotimac was designed for rotifer enrichment for larval finfish culture and is not a pure Schizochytrium diet; it has a higher carbohydrate component and lower lipid fraction (Table 1). 11

27 For this experiment six discrete upwelling systems were used each consisting of a semicircular 75-L reservoir tank (65 em x 45 em x 28 em L x W x D) with a submerged aquarium pump (Hagen 402 Powerhead) pumping seawater into a 135-L upper tank (57 em x 39.5 em x 61 em) fitted with three 55 em x 12 em ID round upwellers with I 00-~-tm nitex screen. Three hundred millilitres of juvenile C. gigas were introduced into each upweller for a total of 900 ml of Pacific oyster seed per treatment. Due to the need of Pearl Sea Products to improve the growth of poor performing oysters, nutritionallydepressed oysters were provided for use in this experiment. All incoming seawater was filtered to 1-~-tm using 25-~-tm and 1-~-tm cartridge filters plumbed in series and aeration was continually supplied to each reservoir tank. When the oysters were not being fed, fresh seawater was pumped into the upwellers and the tanks acted as a flow-through system. Ambient seawater was used and this resulted in water temperatures similar to local winter marine conditions (Harrison et al. 1983). Temperature was measured using automatic temperature loggers (Onset Computer Corporation Stowaway Tidbit -5 C to +37 C). Temperature was measured in the 100% T-Iso tank, the 1:1 (dw:dw) A1gamac 3050 /T-Iso tank and the starved tank to assess the range of possible temperatures (precision ± 0.2 C). Temperature range varied in each of the three measured tanks due to the different initial temperatures of the introduced seawater containing the feeds and as a 12

28 result of the surrounding air temperature gradually warming the water during the six-hour static feeding cycle. The test oysters were provided by Pearl Sea Products from their Sykes lsland FLUPSY for use in this experiment. lt was indicated by Pearl Sea Products' staff that the oysters were those remaining in the mm size range after several size grading events during the 2001 summer growing season. Therefore, the oysters used for this experiment were determined to be slow-growing and unresponsive to the natural diet present at the Sykes Island farm site. In order to ensure that the oysters' enzymatic systems were given sufficient time to be able to efficiently digest the test diets, the oysters were fed daily with cultured Isochrysis galbana (T-Iso) for a period of three weeks prior to the start of the experiment. For the experiment, the oysters were fed to satiation for a period of six hours per day for 60 days. The experimental period lasted from December 11 1 \ February 14 1 \ The time period of 60 days was chosen because the prevailing time period for juvenile oyster feeding studies is days (Knauer and Southgate 1996; Brown et al. 1998; Soudant et al. 2000; Kmtckey et al. 2002). Brown and McCausland ( 1999) determined that a period of 28 days was sufficient to statistically detect a significant difference among diets and to observe a 20% increase in growth from a control diet, Jsochrysis galbana (clone T-Iso). In the study of Brown and McCausland (1999) the experimental temperature was higher and the oysters were not identified as slow- 13

29 growmg. Therefore, for the purposes of this experiment, a longer feeding period was used to allow a sufficient period of growth in order to enable firmer conclusions to be drawn as to the effectiveness of the test diets. ' A six-hour static feeding cycle was used because the total volume of water in the tank systems was not large enough in comparison to the amount of oysters in each tank to ensure that the feed would be sufficient to satiate the oysters for longer than six hours. In order to assume that the oysters were fed to satiation, cell particle levels were measured over a 6 hour feeding period using a Coulter Counter (Model T). The 6-hour feeding cycle was also used to simulate a feeding cycle that could be used in a remote location by commercial oyster farmers during an 8-hour work shift. Diet feeding efficiency (Ore) or the percentage difference of suspended food particles at time 0 (C 1 i) and 6 hours postintroduction (C 11 ) was computed: Diets were introduced into the experimental tank systems and evenly distributed by the aeration of the reservoir tank and the incoming seawater was shut off to create a recirculating upweller system. Following addition of the live algae to each of the mixed diet and live algae control diet tanks, the water level was brought to maximum recirculated volume for all of the tanks by addition of fresh l )..Un-filtered seawater. Live algae substitute diets were assumed to remain suspended in the water column for the 6 hour static feeding cycle. Gouthro eta!. ( 1998) found that there was no settlement of 14

30 Schizochytrium cells out of the water column after 5 hours in a tank system with constant aeration. Dissolved oxygen and ammonia levels were monitored for the first week of feeding to assess the range of values during the six-hour static feeding cycle. Prior to the start of the experiment, 0.2 g of both Schizochytrium-based diets (Algamac 3050 and Rotimac ) were added to 1 L of 1 J.!m-filtered seawater and blended for 3 minutes at high speed in a Waring blender. Number of cells ml- 1 and range of cell sizes were measured using a Coulter Counter (Multisizer II). Algamac 3050 yielded approximately 72 million cells ml- 1 and cells ranged in size from J...Lm. Rotimac yielded approximately 120 million cells ml- 1 and cells ranged in size from f.!m Summary of diets tested For this experiment six treatments were used. A control diet of lsochrysis gatbana (clone T-Iso) was used because it was shown to be an effective diet for juvenile C. gigas in several past studies (Wilson 1987; Brown et at. 1998; Onal and Langdon 1998; McCausland et at. 1999; Brown and McCausland 2000) and it is also in wide use in commercial rearing facilities due to its ease of growing and favourable nutritional profile (Coutteau and Sorgeloos 1992). A starved control treatment was used to assess the ability of the juvenile C. gigas to withstand a prolonged period of starvation by maintaining the oysters in 1 ~tm filtered seawater. The starvation treatment was also used to measure the survival of the nutritionally depressed oyster seed provided by Pearl Sea Products for this experiment and compare it to the other treatments. 15

31 , ~ The two Schizochytrium-based diets that were used as live algae substitute diets were Algamac 3050 and Rotimac supplied by Aquafauna Biornarine (Hawthorne, CA, USA). Algamac 3050 is a pure spray-dried, heterotrophically-grown Schizochytrium diet. Schizochytrium is a thraustochytrid protist species that is naturally occurring in tropical mangrove marine habitats (Raghukumar 1992). Rotimac is a blended Schizochytrium diet that was originally developed for rotifer enhancement due to its high carbohydrate content and much lower lipid content when compared to the Algamac 3050 pure Schizochytrium diet. Table l summarises the biochemical properties of the five test diets. Diets were prepared by mixing a pre-weighed amount in ambient, 1 ).lidfiltered seawater for approximately 3 minutes. A total of six treatments were used in this experiment: a 100% Algamac 3050 diet; a 100% Rotimac diet; a 1:1 (dw:dw) mixed T-Iso/Algamac 3050 diet; a 1:1 (dw:dw) mixed T-lso/Rotimac diet; a 100% T-Iso control diet; and a starved treatment. All diets were measured on a dry weight basis with moisture content of the live algae substitute diets determined prior to the commencement of the experiment. The feeding ration was determined as 1% of the initial dry weight of the oysters. The dry weight of the oysters was calculated using the average moisture content value of three sub-samples taken prior to the start of the experiment. The initial volume of oysters in each upweller was converted to a dry weight equivalent and the dry weight of the ration was determined. For determination of the live algae ration size, a known volume of algae 16

32 was vacuum-filtered with 8 mm diameter glass-fibre filters (pre-combusted at 500 C for 48 hours) and the average dry weight per cell was determined by dividing the dry weight per ml by the mean number of cells per ml in the original sample determined using a Coulter Counter (Model T). Therefore, the live algae ration was determined by volume using daily cell counts Algal culture Isochrysis galbana (T-lso, clone CCMP 1324, Bigelow Laboratories) was maintained as a stock culture by transferring 50 ml of actively growing 48-hour stock culture to 600 ml 20 C pasteurised seawater (1-J..lm and charcoal filtered prior to 85 C pasteurisation) enriched with modified Harrison nutrient mix (Harrison et al. 1980). The Harrison media was modified by substitution of25 g NaH2P0 4 6H20 for 25 g Na2 glycero P0 4 6H 2 0. The remaining culture was transferred to a 20-L carboy containing pasteurised seawater enriched with nutrient mix and was gently aerated. When the cell density reached mid-exponential level, the content of the carboy was transferred to 250-L of pasteurised seawater in a 350-L fibreglass column with nutrient mix added. The culture was gently aerated and ph was kept in the range by bubbling C02 into the culture using solenoids to control the flow when the ph fell below 7.4. A ph meter was installed in the column to regulate C02. All cultures received 24 hour illumination at 250 J.lmol m- 2 s- 1 provided by full spectrum fluorescent lights. The temperature in the culture laboratory was kept constant at 18 ± I oc (L. Keddy, pers. comm.). 17

33 All phytoplankton used in this experiment were harvested during the late exponential phase of development. Cell counts were perfom1ed daily using a Coulter Counter (Model T) Biochemical analysis of oysters and test diets Two oyster samples of approximately 1.5 g wet weight were taken every 10 days from each upweller (yielding triplicate samples per diet) immediately before feeding. The oysters were starved for a period of at least 16 hours prior to sampling in order to allow for clearing of the gut contents prior to proximate analysis. Whole oyster samples were rinsed with cold 0.5 N ammonium formate to remove excess salt, patted dry with a paper towel to absorb excess water and then weighed to determine wet weight. Next, the samples were stored in sealed glass vials and placed at -80 C for 24 hours, lyophilised in a freeze-drying unit for 48 hours and re-weighed to determine dry weight and to calculate original moisture content. All samples were stored as lyophilised samples, sealed in glass vials, vacuum-sealed in a plastic bag, stored at -80 C and were analysed within 8 months of sampling. It was previously shown by Whyte (1987) that samples stored in this manner do not show measurable degradation of biochemical components, notably lipid and protein components following a period of one year. The two separate samples were used for measurements and determination of biochemical components. Survival was estimated by counting the amount of empty bottom valves 18

34 present in the measurement sample (dead oysters were detennined by gently pushing on the umbo and those shells where the valves separated easily and no organic tissue remained inside were counted as dead) versus live oysters. A sub-sample of thirty live oysters was used to measure shell height, width and depth using digital calipers to± 0.02 mm (Mitutoyo Model #CD-8"CS). The biochemical component sample, following the removal of dead oysters, was manually ground to a fine powder using a mortar and pestle. This powdered sample was then used in analyses of ash content, crude protein, lipid, mono- and oligosaccharide, polysaccharide and fatty acid content. To determine ash content and ash free dry weight (AFDW), triplicate samples of approximately 100 mg were weighed into pre-com busted (500 C for 24 hours and stored in a desiccator) porcelain crucibles with aluminium foil covering the opening to prevent loss of sample from explosive combustion. Samples were com busted at 500 C for 48 hours and placed in a desiccator after cooling to 100 C. Samples were weighed to within 0.01 mg after cooling to room temperature on a microbalance (Mettler AT20 1 ). Test diets were lyophilised to determine ash free dry weight and stored at -80 C for a maximum of 8 months prior to analysis. T-Iso samples were taken at Days 0, 30, and 60 by centrifugation with a Sharples centrifuge at 12,000 rpm that provided a paste that was transferred to a plastic bag, frozen to -80 C, lyophilised, and stored at -80 C until analysed. 19

35 Crude protein, lipid, mono- and oligosaccharide, and polysaccharide concentrations were determined using the methodologies described by Whyte (1987). The following energy equivalents were used in calculating the energy potential oflipid, protein and carbohydrate: 9.8, 4.8 and 4.2 kcal g- 1, respectively (Whyte 1987) Statistical analysis Data from the feeding experiment were compared using analysis of variance (ANOVA) followed by Tukey's post hoc test for pair wise comparisons. Prior to analysis by ANOVA, the homogeneity of variance among the upwellers in each tank was confirmed using the Levine test. Data from the upwellers were grouped to yield a single mean value per tank for comparison among tanks. Tamhane's test was used to analyse data that failed the Levine test for homogeneity of variance. The instantaneous growth rate (k) was calculated as outlined by McCausland and Brown (1999) for comparison to past feeding studies and to allow comparison of the relative effectiveness of the reference diet, I. galbana (clone T-Iso) at suboptimal experimental temperatures. k = (lnm 1 - lnmo)lt where M = measured weight (mg) and t = experiment length (days) All results were reported to ± one standard error(± SE). 20

36 Experiment Experimental setup Please refer to section for details on the treatments used for Experiment 2. Samples were taken at Days 0, 20 and 60 for comparison of protease activity. The Day 20 oysters were used to indicate if zymography could be used for early detection of a poor response to a test diet. The original goal of the starvation trial was to measure the loss in nutritional components of the starved oysters to detennine which components were preferentially used during starvation. This would allow one to hypothesize about the relative importance of different nutritional components to juvenile C. gigas and aid in determining the structure of an optimum diet. Due to time constraints and logistical problems, the starvation samples were not processed. During sampling, 15 live oysters were randomly chosen from one upweller in each tank to yield 5 samples per upweller and the oysters were rinsed with cold distilled water to remove faeces and other organic matter that could contaminate the preserved sample. The oysters were then rinsed with cold 0.5 N ammonium formate to remove salt deposits and then dissected to remove all tissue for storage. Throughout sampling, the oysters were kept on ice unless otherwise noted. A total of three individual oysters was grouped to yield a sufficient sample size (>l mg wet weight of tissue) for analysis and also to 21

37 limit potentially confounding genetic variability. Plastic 5 ml genetic storage cryovials were used for storage and all samples were frozen at -80 C for 24 hours, then lyophilised for a minimum of 48 hours, vacuum-sealed in a plastic bag and stored at -80 C until analysed. All samples were analysed within 6 months of sampling Analysis of protease activity Qualitative protease analysis was performed at the Laboratory of Dr. Neil Ross at the National Research Council of Canada Institute for Marine Biosciences in Halifax, Nova Scotia, Canada. Prior to qualitative protease analysis, protein concentration was determined using the Bradford protein assay (Bradford 1976) using a commercial reagent (BioRad). A 250 f.ll aliquot of Bradford Reagent diluted with double distilled water (1 :4) was added to 5 f.ll of unknown sample dissolved in 40 mm NaH 2 P0 4 buffer solution (ph 6.2). Samples were analysed in 5 ml wells (64 well plate) and absorbance at 590 nm read using a Thermomax microplate spectrophotometer (Molecular Devices, Sunnyvale, CA, USA). Bovine gamma globulin was used as a standard. Qualitative protease analysis was performed by separating the protease enzymes using gel electrophoresis (Ross eta!. 2000). Separating gels were formulated by mixing ml double distilled water (ddh 2 0, ml 1.5M TrisHC1 buffer (ph 8.8), 50 f.!l SDS ( 10% w/v) and 2.0 ml 30% Acrylamide/Bis. After degassing for 15 minutes, 0.5 ml 22

38 gelatine (50 mg gelatine dissolved in 5 ml boiling ddh20), 25 JlL 10% ammonium persulfate (APS) and 2.5 JlL TEMED were added. Stacking gels were fonnulated by mixing 3.05 ml ddh20, 1.25 ml 0.5M TrisHCl buffer (ph 6.8), 50 JlL 10% SDS (w/v) and 0.65 ml 30% Acrylamide/Bis. After degassing for 15 minutes, 25 JlL 10% APS and 5 ~-tl TEMED were added. For each tank, a total of five replicates were run for Days 0 and 60. Five 1-1g of sample protein was added to each well and feed samples were also run but in triplicate. Gels were run at 150V until the dye front had reached approximately 25 mm from the bottom of the plate. Gels were then rinsed in wash buffer for 10 minutes three times on ice and then incubated at 30 C for approximately 19 hours in a temperature controlled agitator ( ~ 70 rpm). Following incubation, the gels were stained with 0.1% ami do black in MeOH/H20 /AcOH (45:45:10) for 1 hand destained with MeOH/H20 /AcOH (50:48:2) until desired contrast was obtained Analysis of zymography gels Zymography gels were saved as digitised images and the resultant data were a qualitative assessment of the absence/presence of protease enzymes in the oysters and no quantitative measure of protease activity was attempted. Digitised images of zymography gels were used to identify protease protein bands and low molecular weight protein markers were used to estimate the size of the pro teases. 23

39 2.3 Experiment Experimental setup This experiment was conducted at Unique Seafarms Inc., a commercial scale multispecies bivalve hatchery located in Cedar, British Columbia ( ' N, ' W). For this experiment, three discrete recirculating feeding systems were used, each consisting of a modified circular 165-L reservoir tank ( 106 em x 58 em TD; Barr Plastics, Abbotsford, British Columbia, Canada) connected to an upper 135-L tank (57 em x 39.5 em x 61 em) fitted with three round upwellers (55 em x 12 em ID) with m nitex screen to retain the oysters. A water pump (Iwaki PM20) was plumbed in series to pump water from the bottom of the reservoir tank to the upper tank. All incoming seawater was filtered through two sand filters, then a UV-filtration system and was assumed to be free of particles larger than 1 11m in size. Bloom tank water was introduced into the tanks without filtration. Aeration was provided by the water returning from the upper tank into the reservoir tank. When the oysters were not being fed, fresh seawater was pumped into the upwellers and the tanks acted as a flow-through system. Both the bloom tank seawater and the incoming fresh seawater were at ambient temperature levels and were unheated prior to introduction into the feeding systems. Ambient seawater was used and this resulted in water temperatures similar to local winter marine conditions (Harrison et al. 1983). Temperature (±0.3 C) was measured using automatic temperature loggers (Vemco Incorporated Minilog TX 8-Bit -5 C + 35 C). Temperature was measured in all of the tanks to assess any difference in 24

40 temperature between the mixed diets and the 100% bloom tank algae. Temperature range varied in each of the three measured tanks due to the different initial temperatures of the introduced seawater containing the feeds and as a result of the surrounding air temperature gradually watming the water during the 24-hour static feeding cycle. The test oysters were provided by Unique Seafarrns from their nursery surplus for use in this experiment. The test oysters were oysters from the 2002 growing season that were unsold and had been unfed for most of the winter period. They were of a poor overall quality in comparison to other faster growing seed from the same year class. Therefore, the oysters used for this experiment were detennined to be of a poor overall quality and unresponsive to the bloom tank diet. Unlike Experiment 1, the oysters were not fed prior to the beginning of the experiment and they were placed in the experimental tanks for 48 hours in ambient running seawater to acclimatise before commencement of the feeding trial. As in Experiment 1, 300 ml of oysters were added to each upweller to give a total of 900 ml of oysters per tank. Diets were fed as 1% of the initial dry weight per tank of oysters calculated prior to the start of the experiment by lyophilising three subsamples (~30 individual oysters) of juvenile oysters and using the moisture content to detem1ine the initial oyster dry weight per tank. 25

41 The oysters were batch fed to satiation levels ( l% initial dry weight of oysters) once per day and the tanks were left to recirculate for approximately 24 hours for 30 days. The oysters were assumed to be satiated following the feeding period because food particles were still present. Therefore, the tanks were completely drained and rinsed with cold filtered seawater before addition of the diets. The experiment lasted from January February 23rd, The 30 day time period was chosen strictly for logistical reasons as a longer experiment would not enable completion by August The 24-hour feeding cycle was also chosen for logistical reasons as the after hours access required for a shorter feeding time was not possible at Unique Seafarms Inc.'s facility Summary of diets tested The two Schizochytrium-based diets that were used as live algae substitute diets were Algamac 2050 and Rotimac supplied by Aquafauna Biomarine (Hawthorne, CA, USA). Diets were prepared by mixing a pre-weighed amount in ambient sand- and UVfiltered seawater for approximately 3 minutes. Three diet formulations were tested: a 1: I ( dw:dw) formulation of Rotimac and Thalassiosira nordenskioeldii, a 1: I (dw:dw) formulation of Algamac 2050 and T nordenskioeldii and a 100% T nordenskioeldii formulation. Table 2 shows the proximate component values for each of the pure diets as well as theoretical compositions of the mixed diets based on equal contributions. Table 3 outlines the fatty acid values for each of the pure diets as well as the two mixed diets. Myristic acid 26

42 (14:0), palmitic acid (16:0) and docosahexaenoic acid (DHA; 22:6m3) fatty acid values in the Algamac 2050 and Rotimac diets were much higher than the T nordenskioeldii diet; however, eicosapentaenoic acid (EPA; 20:5m3) was higher in the T nordenskioeldii diet (Table 3). All diets were measured on a dry weight basis with moisture content of the live algae substitute diets determined prior to the commencement of the experiment. The feeding ration was determined as 1% of the initial dry weight of the oysters. The dry weight of the oysters was calculated using the average moisture content value of three sub-samples taken prior to the start of the experiment. The initial volume of oysters in each upweller was converted to a dry weight equivalent and the dry weight of the ration was determined. For determination of the live algae ration size, 50 ml of live algae was filtered through a pre-weighed, glass-fibre filter (combusted at 500 C for 48 hours) that was lyophilised and reweighed to determine dry weight per ml. This estimate was used to calculate the ration size in litres added to the tanks. In the mixed diet tanks, after addition of the bloom tank algae ration, fresh seawater was added to make the final tank volume double that in the 100% T nordenskioeldii tank. Dissolved oxygen was measured daily and ammonia was measured once per day for the first week of the experiment and ammonia was assumed to be below detectable levels for the remainder of the experiment. 27

43 2.3.3 Algal culture Bloom tank algae was grown in outdoor 2000-L concrete tanks supplied with inorganic fertilisers and organic waste by-products provided by recirculating seawater from bivalve rearing tanks. A series of four 2000-L concrete tanks with 2.7 em e/ 4 " ) diameter PVC outlets were plumbed in parallel to a junction with an electric pump that was hooked up to a sterilised rubber hose to deliver bloom tank algae for use in the experiment (rinsed with ~8 0 C freshwater prior to use). Natural sunlight and ambient outside temperature were used for promoting phytoplankton growth in the bloom tanks. Daily phytoplankton samples were taken to assess the dry weight per ml of bloom tank algae. Samples for determination of chlorophyll!were also taken every 3 days. Every 7 days from the commencement of the experiment, 40 L of bloom tank algae was concentrated by centrifugation with a Sharples centrifuge at 12,000 rpm, providing a paste that was transferred to a plastic bag, frozen to -80 C, lyophilised, and stored at -80 C until analysed Biochemical analysis of oysters and test diets Two oyster samples of approximately 1.5 g wet weight were taken every 6 days from each upweller (yielding triplicate samples per diet) immediately before feeding. The oysters were then placed in filtered flowing seawater for 24 hours to allow for clearing of the gut contents prior to analysis of the biochemical components. Samples were rinsed with cold 0.5 N ammonium formate to remove excess salt, patted dry with a paper towel 28

44 to absorb excess water and then weighed to determine wet weight. Next, the samples were stored in sealed plastic vials and placed at -80 C for 24 hours, lyophilised in a freeze-drying unit for 48 hours and re-weighed to determine dry weight and to calculate original moisture content. All samples were sealed in plastic vials, vacuum-sealed in a plastic bag, stored at -80 C and were analysed within 6 months of sampling. The two separate samples were used for measurements and determination of biochemical components. Survival was estimated by counting the amount of empty bottom valves present in the measurement sample (dead oysters were determined by gently pushing on the umbo and those shells where the valves separated easily and no organic tissue remained inside were counted as dead) versus live oysters. A sub-sample of 30 live oysters was used to measure shell height, width and depth (±0.02 mm) using digital calipers to (Mitutoyo Model #CD-8"CS). The biochemical component sample, following the removal of dead oysters, was manually ground to a fine powder using a mortar and pestle. This powdered sample was then used in analyses of ash content, crude protein, lipid, mono- and oligosaccharide, polysaccharide and fatty acids. To determine ash content and ash free dry weight (AFDW), triplicate samples of approximately 100 mg were weighed into pre-com busted (500 C for 24 hours and stored in a desiccator) porcelain crucibles with aluminium foil covering the opening to prevent loss of sample from explosive combustion. Samples were combusted at 500 C for 48 29

45 hours and placed in a desiccator after cooling to 1 00 C. Samples were weighed to within 0.0 l mg after cooling to room temperature on a microbalance (model Mettler A T20 I). Test diets were lyophilised to determine ash free dry weight and stored at -80 C for a maximum of 8 months prior to analysis. Bloom tank samples were taken every 7 days by centrifugation with a Sharples centrifuge at 12,000 rpm. The centrifugation provided a paste that was transferred to a plastic bag, frozen to -80 C, lyophilised, and stored at -80 C until analysed. Crude protein, lipid, mono- and oligosaccharide, and polysaccharide concentrations were determined using the methodologies described by Whyte (1987). Fatty acid concentrations were determined using the methodology described by Whyte (1988) Statistical analysis See section for a description of the statistical methods used for analysis. All results were reported to ± one standard error (± SE). 30

46 3.0 Results 3.1 Experiment Efficiency of diet uptake over a 6 hour feeding cycle The diet feeding efficiency ranged in value from 25.14% (T-Iso) % (Rotimac /T-Iso) (Figure 2). The two mixed diets (Algamac 3050 /T-Iso and Rotimac /T -Iso) yielded the highest percentage of ingested cells (assuming insignificant settlement of cells) at 81.05% for the Algamac 3050 /T-Iso diet and 88.58% for the Rotimac /T-Iso diet (Figure 2). Statistical analysis was not performed on the data as there were only single samples analysed for each diet and time period Experimental temperature The mean temperature in the Algamac 3050 /T-Iso tank was 9.49 ± 0.01 C (range C). The mean temperature in the T-Iso tank was 9.59 ± 0.01 C (range C). The mean temperature in the starved control tank was 9.07 ± 0.01 C (range C) Shell height, dry weight and ash-free dry weight The initial shell height for test oysters (Crassostrea gigas) was 9.59 ± 0.11 mm (± SE). There were no significant differences in shell height between Day 60 oysters and Day 0 oysters for any of the test diets and the starved control. The range in shell height at Day 60 was 9.31 ± 0.26 mm (Algamac 3050 diet) to ± 0.31 mm (Rotimac /T-Iso diet). 31

47 The initial dry weight per individual test oyster was ± 0.42 mg (± SE). Table 4 shows the mean initial and Day 60 individual oyster dry weight per tank and the percentage change in dry weight. The range in individual oyster dry weight at Day 60 was ± 0.35 mg (starved control) to ± 1.54 mg (Algamac 3050 /T-Iso) (Table 4). Both mixed diets (Algamac 3050 /T-Iso and Rotimac /T-Iso) and the T-Iso diet Day 60 dry weight per oyster were significantly higher than the mono Schizochytrium diets and the starved control (Table 4). Individual dry weight for oysters fed the Algamac 3050 /T-Iso diet was significantly higher than the Rotimac /T-Iso and T-Iso diets at Day 60 and there was no significant difference between the Rotimac /T-Iso and T-Iso diets at Day 60 (Table 4). The instantaneous growth rates (k) for each diet tested were as follows: (AM3050 /T-Iso), (AM3050 ), (Rotimac /T-Iso), (Rotimac ), (T-Iso), (Starved). The ash-free dry weight (AFDW) per oyster at Day 0 was 4.63 ± 0.26% (Table 5). The AFDW values at Day 60 ranged from 9.39 ± 0.58% (T-Iso) to 5.50 ± 0.32% (Algamac 3050 ) (Table 5). AFDW was significantly higher in the Algamac 3050 /T-Iso, T-Iso and starved control diet compared to the Schizochytrium mono-diets (Algamac 3050 and Rotimac ) (Table 5). AFDW in the Rotimac /T-Iso diet was not significantly different from the Algamac 3050 diet (Table 5). 32

48 3.1.4 Crude protein content Crude protein content(%) at Day 60 for oysters fed both mixed diets (Algamac 3050 /T-Iso and Rotimac /T-Iso), as well as the T-Iso-fed oysters, were significantly higher (p<0.05) than the oysters at Day 0 (Figure 4). At Day 60 oysters fed both mixed diets (Algamac 3050 /T-Iso and Rotimac /T-Iso) had significantly higher crude protein content(%) than oysters fed the Schizochytrium-based diets (Algamac 3050 and RotimacTM) and the starved control (Figure 4). The crude protein content(%) for oysters fed the T -Iso diet was significantly higher than oysters fed the Algamac 3050 and the starved control; however there was no significant difference between oysters fed the T-Iso and Rotimac diets (Figure 4). Crude protein content (dw) per oyster at Day 60 ranged from 4.87 ± 0.13 mg (Algamac 3050 /T-Iso) to 1.20 ± 0.02 mg (starved control) (Figure 5). The Algamac 3050 /T Iso-fed oysters had crude protein content (dw) that was significantly higher than all other diets (Figure 5). Oysters fed the Rotimac /T-Iso and T-Iso crude protein content (dw) were not significantly different but they were significantly higher than oysters fed both mono-diets (Algamac 3050 and Rotimac ) and the starved control oysters (Figure 5) Crude lipid content Crude lipid content(%) at Day 60 for oysters fed all diets was significantly higher than the oysters at Day 0 (Figure 6). At Day 60 oysters fed the AM3050 /T-Iso diet had significantly higher crude lipid content(%) than all other diets (Figure 6). 33

49 Crude lipid content (dw) at Day 60 ranged from 0.12 ± 0.01 mg ( tarved control) to 1.40 ± 0.09 mg (Algamac 3050 /T-Iso) (Figure 7). Oysters fed the Algamac 3050 /T-Iso diet had crude lipid content (dw) that was significantly higher than oy ters fed all other diets (Figure 7). Oysters fed all remaining diets were significantly higher than the starved control oysters at Day 60 (Figure 7) Crude mono/oligosaccharide content Oyster crude mono/oligosaccharide content(% and dw) at Day 60 for all diets was not significantly higher than the oysters at Day 0 (Figures 8, 9). Tamhane's test was used to analyse significant differences between tanks at Day 0 and Day 60 as equal variances could not be assumed (Levine's test of homogeneity of variances p < 0.00 I). Oyster crude mono/oligosaccharide content (dw) at Day 60 ranged from 0.06 ± 0.01 mg (starved control) to 0.43 ± 0.08 mg (Aigamac 3050 /T-Iso) (Figure 9) Crude polysaccharide content Oysters fed the Rotimac /T-Iso and T-Iso diets showed a significantly higher crude polysaccharide content at Day 60 compared to Day 0 (Figure l 0). Oysters fed the mixed Rotimac /T-lso crude polysaccharide content were not significantly different from oysters fed the Algamac 3050 /T-Iso diet at Day 60. Oysters fed the T-lso diet had 34

50 significantly higher crude polysaccharide content than all diets with the exception of oysters fed the RotimacTM/T-Iso diet at Day 60 (Figure 10). Oyster crude polysaccharide content (dw) at Day 60 ranged from 0.07 ± 0.01 mg (starved control) to 2.25 ± 0.20 mg (T-Iso) (Figure 11). The Rotimac /T-Iso fed oysters' crude polysaccharide content (dw) was significantly higher than all other diets except for the T-Iso and Algamac /T-Iso diets (Figure 11). Oysters fed the T-Iso diet were significantly higher than all diets except for oysters fed the Rotimac /T-Iso (Figure 11 ). Oysters fed all remaining diets were not significantly higher than the starved control oysters at Day 60 (Figure 11 ). 3.2 Experiment Protease activity The oysters at Day 0 showed a single intense band of pro teases between 53 and 91 kilo Daltons (kda; Figure 12). Oysters fed both mixed diets (Algamac3050 /T-Iso, Rotimac /T-Iso) and the T-Iso reference diet showed higher protease activity as well as a higher diversity of protease molecular weights after 20 and 60 days of feeding (Figures 13-16). There were no protease bands present <53 kda in oysters fed the 100% Schizochytrium-based diets and the starved control diet following 20 and 60 days of feeding (Figure 13-16). At Day 20, oysters fed the two mixed Schizochytrium diets and the T-Iso diet showed two distinct protease bands of approximately 35 and 29 kilo Daltons (Figures 13 and 14). These protease bands were also present in oysters fed the 35

51 mixed and T-Iso diets after 60 days of feeding. The lower molecular weight protease bands (29-35 kda) were only present in oysters fed the diets that produced an effective growth response (Figures 15 and 16). 3.3 Experiment Experimental temperature The mean temperature in the Rotimac!Thalassiosira nordenskioeldii tank was ± 0.01 C (range l1.13 C). The mean temperature in the Algamac 2050 /T nordenskioeldii tank was ± 0.01 C (range C). The mean temperature in the T nordenskioeldii control tank was ± 0.01 C (range C) Survival Oyster survival ranged from 86.5 ± 0.6 % (Rotimac /T nordenskioeldii) to 81.3 ± 1.9 % (T nordenskioeldii) at Day 30. Survival at Day 30 for oysters fed the Algamac 2050 /T nordenskioeldii diet was 86.2 ± 1.9% (±SE). There were no significant differences in survival among oysters fed the test diets at Day Dry weight and ash-free dry weight The initial (Day 0) dry weight per individual test oyster was ± 0.20 mg(figure 17). The range in individual oyster dry weight at Day 30 was ± 0.97 mg (T nordenskioeldii) to ± 0.34 mg (Algamac 2050 /T nordenskioeldii) (Figure 17). The individual dry weight for oysters fed both mixed diets (Algamac 2050 /T 36

52 nordenskioeldii and Rotimac /T nordenskioeldii) was significantly higher than oysters at Day 0 and oysters fed the Day 30 T nordenskioeldii diet (Figure 17). There was no significant difference among the mixed diets at Day 30 and between the T nordenskioeldii diet at Day 0 and Day 30 (Figure 17). The instantaneous growth rate (k) was dai 1 for both mixed diets and dai 1 for the T nordenskioeldii diet. The oyster ash-free dry weight (AFDW) at Day 0 was 4.86 ± 0.17% (Figure 18). The oyster AFDW values at Day 30 ranged from ± 0.63% (Rotimac /T nordenskioeldii) to 9.23 ± 0.22% (T nordenskioeldii) (Figure 18). AFDW among oysters fed all test diets at Day 30 was significantly higher than the Day 0 oysters (Figure 18). There were no significant differences in AFDW among the test diets at Day 30 (Figure 18) Crude protein content Oyster crude protein content(%) was 3.10 ± 0.34 at Day 0 and ranged from ± 0.18 (Algamac 2050 /T nordenskioeldii) to 6.54 ± 0.39 (Rotimac /T nordenskioeldii) at Day 30 (Figure 19). Oyster crude protein content(%) at Day 30 was significantly higher than at Day 0 for all test diets (Figure 19). There was no significant difference among oysters fed each of the diets at Day 30 (Figure 19). Oyster crude protein content (dw) at Day 30 ranged from 2.03 ± 0.12 mg (Rotimac /T nordenskioeldii) to 1.16 ± 0.06 mg (T nordenskioeldii) (Figure 20). All test oysters (all 37

53 diets) at Day 30 had a significantly higher protein content (dw) than at Day 0 (Figure 20). Oyster crude protein content (dw) for both mixed diets (Rotimac /T nordenskioeldii and Algamac 2050 /T nordenskioeldii) were significantly higher than oysters fed the T nordenskioeldii diet at Day 30 (Figure 20). There was no significant difference in oyster crude protein content (dw) between the mixed diets at Day 30 (Figure 20) Crude mono/oligosaccharide content Oyster crude mono/oligosaccharide content(%) at Day 0 was ± (Figure 21 ). Oyster crude mono/oligosaccharide content(%) ranged from ± (Rotimac /T nordenskioeldii) to ± (T nordenskioeldii) at Day 30 (Figure 21 ). Oysters fed the Rotimac /T nordenskioeldii diet had crude mono/oligosaccharide content(%) at Day 30 that was significantly higher than oysters at Day 0 (Figure 21). There were no significant differences among oysters fed each of the diets at Day 30 and no difference between the Day 30 and Day 0 oyster crude mono/oligosaccharide contents (%) for the Algamac 2050 /T nordenskioeldii and the T nordenskioeldii diets (Figure 21 ). Oyster crude mono/oligosaccharide content (dw) at Day 0 was ± mg (Figure 22). Oyster crude mono/oligosaccharide content ( dw) ranged from ± mg (Rotimac /T nordenskioeldii) to ± mg (T nordenskioeldii) at Day 30 38

54 (Figure 22). There were no significant differences among oysters fed the test diets at Day 30 and Day 0 (Figure 22) Crude polysaccharide content Oyster crude polysaccharide content(%) at Day 0 was 0.88 ± 0.02% (Figure 23). Oyster crude polysaccharide content(%) ranged from 2.78 ± 0.21 % (Rotimac /T nordenskioeldii) to 2.22 ± 0.07 % (Algamac 2050 /T nordenskioeldii) at Day 30 (Figure 23). The T nordenskioeldii fed oysters at Day 30 had a significantly higher crude polysaccharide content than at Day 0 (Figure 23). There were no significant differences among oysters fed each of the diets at Day 30 and the crude polysaccharide content(%) of oysters fed the mixed diets (Rotimac /T nordenskioeldii and Algamac 2050 /T nordenskioeldii) was not significantly different from the Day 0 value (Figure 23). Oyster crude polysaccharide content (mg) at Day 0 was 0.14 ± 0.01 (Figure 24). Oyster crude polysaccharide content (mg) ranged from 0.86 ± 0.06 mg (Rotimac /T nordenskioeldii) to 0.47 ± 0.02 (T nordenskioeldii) at Day 30 (Figure 24). There were no significant differences in crude polysaccharide content (mg) among oysters fed the test diets at Day 30 and among oysters fed the test diets at Day 30 and the Day 0 value (Figure 24). 39

55 3.3.7 Crude lipid content Crude lipid content(%) at Day 0 was 0.31 ± 0.01 % (Figure 25). Crude lipid content (%)ranged from 1.27 ± 0.11 % (Algamac 2050 /T nordenskioeldii) to 0.73 ± 0.04% (T nordenskioeldii) at Day 30 (Figure 25). There were no significant differences among oysters fed each of the diets at Day 30 or Day 0 (Figure 25). Oyster crude lipid content (mg) at Day 0 was 0.06 ± 0.01 (Figure 26). Oyster crude lipid content (mg) ranged from 0.94 ± 0.09 (Algamac 2050 /T nordenskioeldii) to 0.13 ± 0.01 (Rotimac /T nordenskioeldii) at Day 30 (Figure 26). There were no significant differences in crude lipid content (mg) of oysters fed each of the diets at Day 30 and Day 0; however, the crude lipid content of oysters fed the T nordenskioeldii diet was significantly higher at Day 30 than oysters fed the Rotimac /T. nordenskioeldii diet at Day 30 (Figure 26) Fatty acid content Table 6 shows the fatty acid content (mg g- 1 dry weight and% total fatty acid) for juvenile oysters at Day 0 and at Day 30. Myristic acid (14:0) content(% TFA) increased approximately 200% in the oysters fed the Algamac 2050 mixed diet and there was little increase in myristic acid (14:0) in the oysters fed the Rotimac mixed diet and the bloom tank algae diet (Table 6). Palmitic acid (16:0) content (% TF A) in the oysters following 30 days offeeding showed a similar increase (~ 100% ) in the oysters fed the lipid-rich Algamac 2050 mixed diet, as compared to the Rotimac mixed diet and the 40

56 bloom tank algae diet (Table 6). Eicosapentaenoic acid (20:5n3) content(% TFA) increased ~ 100% in the oysters fed the bloom tank algae (comprised mainly of the diatom species Thalassiosira nordenskioeldii). Docosahexaenoic acid (22:6n3) content (% TF A) showed the largest increase in the oysters fed the Algamac 2050 mixed diet (~ 140%). 41

57 4.0 Discussion The mixed diets tested as partial replacements for a live phytoplankton diet produced identical growth characteristics to the I. galbana (clone T-Iso) reference diet in the case of the Rotimac /T-Iso diet (k = day- 1 ) and superior growth with respect to the Algamac 3050 /T-Iso diet (k = day- 1 ). This indicated that both Algamac 3050 (AM3050 ) and Rotimac (RM ) were effective as lower-cost partial replacements for a live phytoplankton diet in a hatchery or nursery setting. Complete replacement of live microalgae by AM3050 and RM were also tested and both diets failed to produce growth. Growth characteristics for the 100% Schizochytrium-based diets were not significantly different than the starved control diet following 60 days of feeding. The pre-experiment 1 measurement of suspended particles during a six hour feeding cycle demonstrated that cells were being actively removed from the recirculating seawater. There was active filter feeding of the suspended AM3050 and RM particles but it is likely that they were rejected as pseudofaeces. Both Boeing (1997) and Brown and McCausland (2000) concluded that a similar Schizochytrium-based diet (AM2000 ) was ineffective as a single diet for feeding juvenile Crassostrea gigas. Experiment l showed that AM3050 and RM are also ineffective as single diets. Schizochytrium was shown to be effective as single diet for feeding Mytilus galloprovincialis so it is likely that C. gigas lacks the necessary enzymes to derive sufficient nutrition from Schizochytrium cells (Whyte et al. 2003). 42

58 Comparison to past studies is difficult given the lower experimental temperature used in Experiments 1 and 3. Both feeding experiments were undertaken during the northern temperate winter season (between November and April). This resulted in experimental seawater temperatures that were not optimal for achieving maximum growth potential. The low experimental seawater temperatures (9 - l 0 C) used for both feeding experiments caused a lack of a comparable growth response to previous studies that used higher experimental temperatures. There are no prior feeding studies involving C. gigas that used temperatures lower than 14 C (Langdon and Waldcock 1981; Waldcock and Holland 1984; Parker and Selivonchick 1986; Knauer and Southgate 1996; Thompson et al. 1996; Knauer and Southgate 1997; Brown et al. 1998; Caers eta!. 2000; Robert eta!. 2001; Flores-Vergara eta!. 2004). Enzymatic systems function more efficiently with rising temperature until excessively high temperatures cause damage to the enzyme's protein structure and limit the effectiveness (Dahlhoff et al. 2002). For molluscs, there is a linear relationship between the rate of enzymatic activity and temperature with oxygen consumption increasing with increased enzymatic activity (Dahlhoff eta!. 2002). Therefore, the low experimental temperatures limited the amount of growth possible during the experimental period because metabolic activity was not operating efficiently. Due to the effect oflow temperature on metabolic rates of ectotherrnic bivalves it is assumed that the growth response would be higher in seawater > 14 C. McCausland et a/. ( 1999) fed juvenile oysters I. galbana (clone T-I so) and achieved an instantaneous growth rate of da/ 1 at temperatures ranging from C, 637% higher than 43

59 the I. galbana (clone T-Iso) fed oysters in the present study (k = day- 1 ; Experiment 1 ). In laboratory studies, the use of higher experimental temperatures is beneficial as it takes advantage of higher metabolic activity and contributes to larger differences in growth parameters such as shell height and dry weight (Flores-Vergara et al. 2004). It also mimics the optimal growing conditions that occur during summer months (May September) when oyster growth is accelerated by high natural phytoplankton abundance and elevated oceanic temperatures. Although this practice will accentuate differences in dietary outcome, it fails to elucidate the effect that differing diets would have on Crassostrea gigas under non-ideal conditions likely to occur outside of the relatively small summer window of maximum growth. The winter period represents a time of the year where food availability is constrained by low seawater temperatures and natural light. It is a period that could greatly enhance growth and survival of existing farm stock if the oysters were able to postpone the winter hibernation-like state and continue feeding. Filter-feeding is accelerated and digestion is faster and more efficient as enzymatic systems function at peak capacity. Faster digestion of incoming nutrients allows more excess energy to be available as base metabolic demands are met relatively quickly allowing surplus energy to be used for anabolic processes or stored for later use. Although the maximum growth rate for Experiment 1 was lower than in past studies using higher temperatures (McCausland et al. 1999; Brown and McCausland 2000), 44

60 when the lower metabolic rate and poor initial physiological condition of the test oysters are considered, the results look promising. The overall condition of the oysters improved during the 60 day feeding cycle and there were increases in all proximate components for both mixed diets and the reference diet. Lower seawater temperature was a factor in the lower growth rate; however, the oysters were previously identified as poor performing seed and grew slower than other oysters in the same cohort under identical growing conditions. Therefore, it is possible that poor genetic characteristics were partially responsible for the lower growth rate. The low initial condition index' (4.63 ± 0.26%) of the test oysters indicated that they were nutritionally depressed with low energy stores. A survey of incoming seed lots for Pearl Sea Products from showed a range in initial condition indices for 2-6 mm juveniles of 6-13 % (Babuin, unpublished data). Although more research is needed to correlate initial condition index with growth and survival, it is assumed that oysters with condition indices <6 % represent individuals with depleted energy stores. The combination of low experimental temperature and poor genetic traits contributed to a minimal change in shell height following 60 days of feeding. There are no previous oyster feeding studies that used slow-growing oysters for dietary testing (Langdon and Waldcock 1981; Waldcock and Holland 1984; Parker and Selivonchick 1986; Knauer and Southgate 1996; Thompson eta!. 1996; Knauer and Southgate 1997; Brown eta!. 1998; McCausland eta!. 1999; Caers eta!. 2000; Robert eta! ). The use of slow- 1 Condition index was calculated asci = ((0/C)*JOO) where 0 = organic content and C = ash content 45

61 growing or poor performing oysters and winter conditions for Experiments 1 and 3 allowed for a rigorous assessment of the effectiveness of supplemental feeding for increasing energy stores prior to the limited C. gigas growing season. The low expetimental temperature resulted in a minimal change in shell height among all diets tested, a measure of dietary effectiveness in some past studies (Holland and Spencer 1973; Holland and Hannant 1974; Wilson 1978; Brown 1988). If only changes in shell height were used to assess dietary effectiveness, the results of both feeding experiments would suggest that all diets tested were ineffective. Change in shell height is a popular measure of dietary effectiveness as it reflects the rate at which the experimental oysters will reach a marketable size as well as being a cost-effective means to assess differences between diets. A prior study (Holland and Spencer 1973) showed that shell height increased while juvenile Ostrea edulis were being starved and it cannot be a useful indicator of dietary effectiveness for shorter duration studies ( <30 days). This is further supported by the resulting 106% increase in mean weight for the Algamac 3050 /T-Iso diet during Experiment 1 coupled with a 7% increase in shell height following 60 days of feeding. Most nutritional studies measure shell height and dry weight as indicators of the dietary performance. Complete proximate analysis represents a more thorough approach to understanding the effect that different diets have on the physical makeup of the oysters and how dietary energy is utilised and stored. Proximate analysis provides a powerful 46

62 tool for assessing dietary effectiveness and determining nutritional requirements for achieving superior growth. The low experimental temperature, poor prior growth characteristics of the test oysters and the limited duration of feeding ( 60 days) necessitated the addition of proximate analysis for determining differences among test diets. There were no significant differences in shell height among all diets tested at Day 60 and Day 0 as was discussed previously. The AM3050 /T-Iso diet produced a significant increase in dry weight (mg) over all other diets tested and the RM /T-Iso and T-Iso diets produced significantly higher average dry weights (mg) than the AM3050, RM and starved control diets at Day 60. The data suggest that live phytoplankton were required to encourage a feeding response or that the AM3050 and RM diets were missing one or more key dietary components necessary to result in growth. The large increase in mean weight compared to the change in shell height indicated that the 60 day feeding cycle represented a period of replenishment in depleted energy stores. Ash free dry weight (AFDW) was significantly higher in the AM3050 /T-Iso and T-Iso diets compared to the AM3050, RM and starved control diets. The subsequent increase in AFDW and condition index supported the conclusion that depleted energy stores were first replenished prior to having excess energy for resuming growth. This lag period in the resumption of growth was also influenced by the low temperature. The oysters' metabolic rate was negatively affected by the low experimental temperature and it is likely that the oysters were not able to assimilate all available dietary energy despite the 47

63 satiation levels of food particles present. Further research is required to assess the effect of low temperature on growth rates and dietary utilisation, specifically, the efficiency of enzymatic function in C. gigas and digestibility of all available dietary components at suboptimal temperatures. Oyster protein content (mg dw) was significantly higher in the AM3050 /T-Iso diet than oysters fed all other diets and this is not surprising given the > 100% increase in oyster dry weight (mg) during feeding (Figure 5). The higher oyster protein content of the AM3050 /T-Iso diet represented a 290% increase over the initial Day 0 oyster protein content (Figure 5). The RM /T-Iso and the T-[so-fed oysters' protein contents (mg dw) were significantly higher than oysters fed the AM3050, RM and starved control diets. The significantly higher protein content of oysters fed the AM3050 /T Iso diet indicated that the oysters were utilising dietary protein from the Schizochy trium cells. If the oysters were not able to use the protein present in the Schizochytrium cells, there would be a comparable increase in protein content to the T-Iso diet or a smaller amount as the mixed AM3050 /T-Iso diet presented a lower density oft-iso cells. The RM /T-Iso and T-Iso diets both had higher crude protein content than the AM3050 diet yet produced a smaller increase in protein content (mg dw). There were also no significant differences in protein content(%) among oysters fed these three diets (Figure 4). This indicated that there were other factors that contributed to the better performance of the AM3050 /T-Iso diet. It is likely that all of the protein assimilated by the oysters 48

64 fed the mixed diets was derived from the T-Iso portion of the diet because the oysters were fed to satiation level. An interesting result was the increase (insignificant) in protein content of the oysters fed the RM diet. The higher carbohydrate content of the RM diet provided more polysaccharide and monosaccharide for utilisation as energy thus allowing the oysters to conserve protein rather than utilise it as an energy source. This was evident in the decrease in protein content of oysters fed both the AM3050 diet and the starved control diet (Figure 5). C. gigas adults were shown in a past study to utilise protein as a source of energy during extended periods of starvation (Whyte 1994) and the results of Experiment 1 indicate that protein was used during 60 days of starvation by the AM3050 and starved oysters. An organism cannot grow if protein is not being assimilated from dietary sources and the data from Experiment 2 confirmed this assumption by measuring protease activity in the starved and fed oysters. Pro teases are enzymes that aid in digestion of dietary protein and the presence of proteases may be used to assess the metabolic activity of an organism similar to the use ofdna:rna ratios as an indicator of metabolic activity (Wright and Hetzel 1985). If protease enzyme molecules are not present, it can be assumed that the test oysters were not assimilating dietary protein and building tissue. Since the structural tissues in oysters are mainly comprised of protein, protease activity was used to determine growth activity. In addition to the evidence of a lack of assimilation of nutritional components from the 49

65 Schizochytrium-based diets by the test C. gigas juveniles, the results from the analysis of protease activity (Experiment 2) also demonstrate a lack of measureable protease levels in starved oysters and oysters fed mono-schizochytrium-based diets at Day 20 and Day 60 (Figures 13-16). Proteases in the molecular weight range of29-35 kda were only present in oysters that were fed superior diets (Figures 13-16). Therefore, the data suggest that proteases were secreted by the test oysters when presented with a suitable diet indicating that digestion and assimilation of dietary protein was taking place. The lack of protease activity in oysters fed both the starved control and the Schizochy triumbased diets suggests an indirect correlation between protease levels and weight gain (Figures 3 and 13-16). The use of protease measurements was shown in Experiment 2 to be an effective means of assessing dietary suitability and performance and may serve as an indicator of growth potential of oysters on a particular diet. Further experimentation is needed to assess whether this technique could be used for testing a diet's acceptance in a time period shorter than 30 days. If an effective and replicable protocol was developed, it would have the potential to make diet evaluation much more efficient by eliminating the need for time-consuming proximate analysis and could also save money. Due to the qualitative nature of the zymography used to assess protease activity, it was not possible to assess the overall rate of protease activity for each of the test diets. Other protease assays could be used to quantify protease levels as a function of diets. It was 50

66 clear from the statistically insignificant differences in mean dry weight (mg) among both 100% Schizochy trium-based diets and the starved control that the C. gigas juveniles were able to enter a hibernation-like state of low metabolic activity and stored energy was mostly conserved for 60 days (Figure 3, Table 4). Figure 3 shows an increase in mean weight for oysters fed both mixed diets (AM3050 /T-Iso and RM /T-Iso) and the 100% T-Iso diet following 40 days of feeding. This suggests the oysters exited a period of hibernation and low enzymatic activity and were able to more efficiently assimilate available dietary components. When C. gigas are presented with nutritionally inferior or foreign food particles, they enter a state of aestivation to conserve energy stores until a more suitable diet is available (J.N.C. Whyte, pers. comm.). The ability to enter an energy conserving state would allow oysters grown in southern BC waters to withstand the extended periods between major algal blooms when natural food abundance is low. Lipid content(% and dw) was significantly higher in oysters fed the AM3050 /T-Iso diet than all other diets (Figures 6 and 7). The lipid content(%) of oysters fed the AM3050 diet was 239% higher than the oysters fed T-Iso and it resulted in much higher crude lipid levels (% and mg dw) (Figures 6, 7). All of the remaining diets produced lipid contents(% and mg dw) that were higher than the starved control oysters at Day 60 (Figures 6 and 7). Therefore, it can also be assumed that the Schizochy triumderived dietary lipid was being metabolised by the juvenile C. gigas because lipid content was not significantly different among all single diet treatments (T-Iso, AM3050, RM ). The high content of total lipid present in the AM3050 diet 51

67 appeared to be effectively assimilated when presented with a high quality phytoplankton diet. There are several past studies that have focussed on total lipid and key fatty acids as being the most important dietary components when evaluating test diets (Langdon and Waldock 1981; Thompson eta!. 1996; Knauer and Southgate 1997; Caers eta!. 2000; Pemet 2005; Milke eta!. 2006, 2008; Garcia eta!. 2008). The increase in lipid content in oysters fed the AM3050 diet indicated that there was ingestion and assimilation of the dietary lipid present in the Schizochytrium cells (Figures 6 and 7). The increase in percent lipid content and dry weight of lipid in the Schizochytrium-based single treatment diets failed to promote growth and the subsequent loss of protein over 60 days indicated that lipid stores were not used as an energy source by the oysters. Dietary lipid present in AM3050 was assimilated and contributed to the higher growth rate seen in oysters fed the mixed AM3050 /T-Iso diet. The higher lipid content of the AM3050 /T-Iso diet provided the oysters with 31% more available energy per unit measure assuming 1:1 inputs of each diet and total oxidation (Table 1). The RM /T-Iso diet and the T-Iso provided similar available energy per unit measure and there were no significant differences among lipid content(% and mg dw) for oysters fed those diets. Fatty acid analysis was not completed for Experiment 1 due to time constraints. Fatty acid data were analysed for Experiment 3 and will be discussed along with the results from that experiment. 52

68 There were no significant differences among oysters fed each of the diets with respect to monosaccharide content(% and mg dw). This was not an unexpected result considering monosaccharide would be utilised for short term energy requirements or converted to glycogen, a polysaccharide for storage. Therefore, major differences in carbohydrate content among oysters fed the differing diets would be expressed in the change in polysaccharide content over time. The Rotimac diet had the highest carbohydrate content of the test diets (35.0%) yet the T-Iso polysaccharide content was not significantly different from the RM /T-Iso diet following 60 days of feeding (Figures 10 and 11 ). This indicated that the dietary carbohydrate present in the RM and AM3050 diets was not being digested. Since T-Iso was shown to support oyster growth, it is likely that the dietary carbohydrate present in the T-I so in the mixed diets was sufficient to meet both short and long term energy requirements. It is possible that the form of the carbohydrate present in the Schizochytrium-based diets was non-digestible by C. gigas and that this caused the lack of a growth response in the oysters fed the AM3050 and RM diets. Further research is required to analyse the carbohydrate present in the Schizochytrium cells and determine if C. gigas is capable of digesting the saccharides present in Schizochytrium. The results of Experiment 1 were consistent with past research involving partial replacement of live phytoplankton by Schizochytrium-based diets (Boeing 1997; Brown and McCausland 2000). Both of those research projects tested an earlier Schizochytrium 53

69 ~ fom1ul ation Algamac This experiment represents the first time that Algamac 3050 and Rotimac were tested for feeding C. gigas. It is clear that Schizochytrium did not stimulate feeding and the results suggest that the carbohydrate present in the Schizochytrium cells is non-digestible for juvenile C. gigas. Oysters fed the AM3050 and RM diets were indistinguishable through proximate analysis from the starved control oysters. The addition of the analysis of protease activity confirmed that there was little digestion of dietary protein taking place (thus indicating a lack of growth activity) and the oysters had entered a hibernation-like state of energy conservation. The low experimental temperature aided in the conservation of energy and it is hypothesised that if the experimental temperature approximated mid-summer conditions (15-20 C) the increase in metabolic rate would result in an energy deficit producing higher mortalities. Further research is needed to confirmed this hypothesis. Both mixed diets (l: 1 dw:dw) provided sufficient nutrients to sustain maintenance energy. The lack of a significant increase in shell height following 60 days of feeding, combined with the poor initial condition of the oysters, indicated that excess energy was being used for rebuilding depleted energy reserves and emaciated tissues. The increase in dry weight (mg) for the mixed diets and the T-Iso diet indicated that the diets provided adequate nutrition with the AM3050 /T-Iso diet providing the highest amount of energy (J mg- 1 ) and also producing the highest increase in dry weight (mg) (Table 4). The mixed RM /T-Iso diet produced results that were very similar to the T-Iso diet; however, RM is less expensive (US$ 20 kg- 1 versus US$ 28 kg- 1 ) than AM3050 and 54

70 it is much less expensive than cultured live algae (estimated US$ 200 kg- 1 : Coutteau and Sorgeloos 1992). AM 3050 was the superior supplemental diet but both Schizochytrium-based diets would result in cost savings for hatcheries, nurseries and grow out operations by increasing storage energy and providing more surplus energy for anabolic processes. A more detailed cost/benefit analysis is required to properly determine the efficiencies of using AM3050 versus RM for supplemental feeding. Experiment 3 tested a different Schizochytrium-based dietary formulation, Algamac 2050 and Rotimac, as partial replacements for bloom tank algae (>95% T nordensh.'ioeldii). Survival(%) was highest in the mixed diets (~86) versus the bloom tank algae diet (~81) but was not significantly different. Both mixed diets (AM2050 /T nordenskioeldii and RM /T nordenskioeldii) resulted in a~ 100% increase in both dry weight (mg) (Figure 17) and AFDW (mg) (Figure 18) compared to the control diet (1 00% T nordenskioeldii) following 30 days of feeding. AFDW was highest in the RM /T nordenskioeldii (11.1 %) and was the highest value of any oysters tested (Experiments 1 and 3). The instantaneous growth rate (k) for both mixed diets (0.022 dai 1 ) was comparable to the growth rate achieved by Brown and McCausland (2000) with juvenile C. gigas fed Algamac 2000 (0.024 dai 1 ) as a supplemental diet for naturally occurring phytoplankton. This result is encouraging when one factors in the higher experimental ocean temperatures (14-19 C) used by Brown and McCausland (2000). 55

71 Oyster protein content (mg) was significantly higher for both mixed diets at Day 30 compared to the control diet and protein content was not a limiting factor in the diets as the T. nordenskioeldii diet had the highest crude protein content (35.5%) yet produced the smallest increase in protein content (135% increase in mg dry wt.: Figure 20). There was no significant difference between protein content(% or mg) for oysters fed the mixed diets despite the protein content of the RM /T nordenskioeldii being ~39% greater than the AM2050 /T nordenskioeldii diet (Figures 19 and 20). Most of the observed difference in protein content (mg dw) between oysters fed the control diet and the mixed diets occurred between day 24 and 30 as the protein content, a measure of overall metabolic activity and growth, of oysters fed the mixed diets increased substantially (Figure 20). Webb and Chu (1983) found that most phytoplankton species that were tested for feeding oysters were able to provide adequate protein and amino acids. The lack of significant difference between the protein content (%) for all diets at day 30 demonstrated that the protein available for all three diets was adequate to promote growth (Figure 19). The dietary protein utilised by the oysters could have been provided exclusively by the T nordenskioeldii portion of the diet because food levels provided were to satiation level. The results of Experiment l agreed with the conclusion that the oysters were preferentially utilising the dietary protein available in the phytoplankton portion of the diet. The RM /T nordenskioeldii diet resulted in significantly higher monosaccharide content (%) following 30 days of feeding (Figure 21 ). This result is not surprising 56

72 considering the monosaccharide level ofthe RM /T nordenskioeldii diet was 46% higher than the AM2050 /T nordenskioeldii diet and 12% higher than the T nordenskioeldii diet (Table 2). The higher level of monosaccharide following 30 days of feeding indicated that the oysters had more simple sugars available for energy conversion. This represented an improvement in the condition of the oysters following 30 days of feeding. The overall level of monosaccharide as a component of the total energy stores of the oysters was small and adequate to meet short term energy requirements. It is assumed that excess monosaccharide present in the RM /T nordenskioeldii diet was converted to glycogen for storage. This was evident in the higher percentage of polysaccharide content in the oysters fed the RM / T nordenskioeldii diet (Figure 23). Polysaccharide content (mg dw) was higher in both mixed diets compared to the T nordensldoeldii diet but there were no significant differences among the oysters following 30 days of feeding (Figure 24). The polysaccharide glycogen is the major energy storage component in the oysters and its assimilation through carbohydrates in food organisms is paramount. Diatoms have higher carbohydrate content than flagellate species and diatoms comprise the majority of phytoplankton species present during the two major annual microalgae blooms in BC (Harrison et al. 1983). The polysaccharide content(%) was significantly higher after 30 days of feeding in the oysters fed the T nordenskioeldii diet despite there being no 57

73 significant differences among oysters fed the three diets at Day 30 (Figure 23). The increase in polysaccharide content for the T nordenskioeldii diet indicated that the carbohydrate supplied by the diatom T nordensldoeldii was the main source of digestible carbohydrate for the test oysters. This result agrees with the previous finding (Experiment l) that the chemical structure of dominant carbohydrates in Schizochytrium cells is unaffected by saccharide enzyme activity in C. gigas and limits carbohydrate assimilation. The AM2050 /T nordenskioeldii diet produced oysters with the highest lipid content (%and mg) but there were no significant differences among oysters fed the diets at Day 0 or Day 30 (Figures 25 and 26) despite the >600% difference in lipid content (mg) between the mixed diets (Figure 26). There were no clear trends with regards to the differences between diets as the total energy available in the mixed AM2050 diet was 15% higher than the T nordenskioeldii diet yet produced an approximate 80% increase in growth (mg). Palmitic acid (16:0) was 189% and 99% higher in the oysters fed AM2050 / T nordenskioeldii diet than the T nordenskioeldii and RM /T nordenskioeldii diets, respectively (Table 6). Eicosapentaenoic acid (EPA; 20:5ro3) was highest in oysters fed the T nordenskioeldii diet, reflective of the approximately 100% higher levels in the T nordenskioeldii (Tables 3 and 6). Docosahexaenoic acid (DHA; 22:6ro3) was 87% and 288% higher in the RM /T nordenskioeldii and AM2050 /T nordenskioeldii diets 58

74 compared to the T nordenskioeldii diet (Table 3). Previous research has shown that EPA and DHA are essential fatty acids for C. gigas (Langdon and Waldock 1981; Waldock and Holland 1984; Thompson et al. 1996; Knauer and Southgate 1997; Caers et al. 2000; Knuckey et al. 2002). The results of this experiment support those findings with respect to DHA but EPA was highest in the T nordenskioeldii diet that produced the least amount of growth (Table 6). Garcia et al. (2008) noted that EPA is a structural component of membranes and can also be oxidised as an energy source. Thompson et a!. (1996) found that an abundance of dietary EPA had a negative effect on growth rates of larval C. gigas. The high retention of DHA in the mixed diets that had the highest increase in dry weight (mg) suggests a major role for this fatty acid in supporting growth. The PUF A docosapentaenoic acid (DPA 22:5n6) was identified in previous feeding studies involving larval sea scallops (Placopecten magellanicus and Argopecten irradians) as a possible essential fatty acid (Pemet et al. 2005; Milke et al. 2006, 2008). The juvenile oysters fed the two mixed diets (RM IT nordenskioeldii and AM2050 /T nordenskioeldii) that demonstrated superior growth to the single T nordenskioeldii diet had DPA levels that increased from trace amount to 4.17% and 6.40% of the total fatty acids respectively (Table 6). Although it is difficult to correlate growth performance to DPA given the similiarity in growth outcome and the varied levels ofdpa in AM2050 and RM, the accumulation ofthis fatty acid in the test oysters demonstrating superior growth warrants further research. 59

75 There was no clear relationship between oyster growth and the proximate composition of each diet, but the structural nature of components would affect this coitelation. Brown et al. ( 1998) tested several phytoplankton diets and also concluded that the growth rates of test C. gigas were correlated with available assimilatable energy and not the gross composition of the individual diets. The energy composition of the single phytoplankton diets was a better indicator of dietary outcome (Brown et al. 1998) because they contained components assimilatable by the oyster and the results of this experiment are in agreement with those findings with respect to the mixed AM2050 diet but not the mixed RM diet. The total energy (J mg- 1 ) assuming total oxidation was 31% higher in the AM2050 diet and 8% higher in the RM diet when compared to the T nordenskioeldii diet (Table 2). Assuming the oysters ingested food particles in the same ratio (I: 1), the total energy (J mg- 1 ) assuming total oxidation was 15% and 4% higher in the mixed AM2050 and RM diets, respectively (Table 2). The energy content of the oysters following 30 days of feeding was approximately l 00% higher in both mixed diets compared to the T nordenskioeldii diet (Figure 27). The use of both AM2050 and RM as 50% replacements for live phytoplankton was more effective than feeding with bloom tank algae (T nordenskioeldii). It is unclear why the mixed Schizochytrium-based diets were superior to the bloom tank algae diet. It is generally accepted that mixed diets are better at providing a balanced diet; species of differing proximate composition can provide higher levels of specific nutrients (Webb and Chu 1983). Schizochytrium-based diets were insufficient as single diets for C. gigas 60

76 juveniles and the results of this experiment suggested that the diatom T nordenskioeldii is also lacking one or more key dietary components, possibly lipid or more specifically the fatty acid DHA (22:6w3). The success of the mixed diets was evident in the higher instantaneous growth rate (0.022 dai') than the mixed AM3050 /T-Iso diet (0.012 dai 1 ) in Experiment 1. The average experimental temperature was ~ 1 C higher in Experiment 3 and the oysters were obtained from different origins and, therefore, genetic differences could have also played a role in the difference in peak growth rate between the two experiments. 61

77 5.0 Conclusions and Recommendations Prior studies used bloom tank algae to enhance oyster production and it resulted in comparable growth to feeding with intensively cultured phytoplankton (Spencer et al. 1986; Laing and Chang 1998). Bloom tank algae can be produced at remote oyster growing locations using inexpensive culture systems and small amounts of commercially available fertilisers. Supplementation of enhanced natural phytoplankton with Schizochytrium-based feeds that are commercially available and can be stored at 4 C for up to one year should enable oyster producers to enhance production in a cost-effective manner. Rotimac was the least expensive Schizochytrium-based product used in both feeding experiments and it produced enhanced growth under suboptimal experimental temperature conditions using previously identified slow-growing C. gigas juveniles. The final proximate composition of the test oysters represented a marked improvement in physical condition and storage energy. The rehabilitation of poorly-performing oysters has the potential to improve time to market, survival and growth efficiency by building up storage energy levels and providing excess energy for fuelling anabolic processes. The original aim ofthis research was to test the effectiveness of supplemental feeding as a means of mitigating summer mortality of C. gigas juveniles. The lack of a significant difference in survival between test oysters for both feeding experiments seemingly contradicted the hypothesis that the summer mortalities were caused by a lack of sufficient nutrition. There are two factors to consider. First, the experimental temperature was lower than the average oceanic temperature during the Northeast Pacific 62

78 summer season (June- September). Secondly, the oysters were only presented with a single component diet consisting mainly of a Schizochytrium strain that is naturally occurring in tropical mangrove habitat, or seawater screened to bacterial level (1 ~-tm). C. gigas are unable to effectively select particulates <2.5 ~-tm in size (Somin eta!. 1988). Therefore, both alternatives presented foreign particulates or no particulates and the oysters were not stimulated to continue to utilise stored energy for feeding activity. It is hypothesised that low levels of nutritious micro algal species present in natural C. gigas growing waters will continue to stimulate feeding activity. Past studies that used chlorophyll content as a measure of overall microalgal abundance showed that low levels of microalgae are consistently present in natural temperate oceanic habitats (Brown and McCausland 2000). Therefore, feeding activity coupled with higher metabolic activity and low natural microalgal availability could result in an energy imbalance that would deplete stored energy. In a juvenile C. gigas individual with low stored energy, this situation could result in critically low maintenance energy levels and possible mortality. A contributing factor would also be higher seasonal oceanic temperatures resulting in increased metabolic activity. Flores-Vergara et al. (2004) experimented with feeding juvenile C. gigas (initial shell height ~ 3 mm) a superior diet(!. galbana (clone T Iso)/Pavlova lutheri) and an inferior diet (Dunaliella tertiolecta) at temperatures ranging from C. They found that at temperatures above 30 C growth was negatively affected regardless of food quality. Oceanic temperatures do not reach 30 C in British Columbian waters and the effect of acclimation to the cooler temperatures prevailing in 63

79 northern temperate oceans must be considered. It is probable that growth could be negatively affected at a lower temperature in British Columbian waters. The extended period of low natural phytoplankton availability between the spring and autumn diatom blooms also coincides with peak oceanic temperatures (~24 C). Further research is required to determine if high experimental temperatures and low food availability of diatom of flagellate species will cause an energy imbalance and mortalities in juvenile C. gigas. 64

80 6.0 References Albentosa, M., A. Perez-Camacho, U. Labarta, M.J. Femandez-Reirez Evaluation of freeze-dried microalgal diets for the seed culture of Ruditapes decussatus using physiological and biochemical parameters. Aquaculture 154: Almeida, M.J., J. Machado, J. Coimbra Growth and biochemical composition of Crassostrea gigas (Thunberg) and Ostrea edulis (Linne) in two estuaries from the north of Portugal. J. Shellfish Res. 18( l ): Alpine, A. E., J.E. Cloem Trophic interactions and direct physical effects control phytoplankton biomass and production in an estuary. Limnol. Oceanogr. 37(5): Baldwin, B.S., R.I.E. Newell Feeding rate responses of oyster larvae (Crassostrea virginica) to seston quality and composition. J. Exp. Mar. Bioi. Ecol. 189: Battaglene, S.C., J.E. Seymour, C. Ramofafia, I. Lane Spawning induction of three tropical sea cucumbers, Holothuria scabra, H fuscogilva and Actinopyga mauritiana. Aquaculture 207:

81 Bayne, B.L., D. Hedgecock, D. McGoldrick, R. Rees. 1999a. Feeding behaviour and metabolic efficiency contribute to growth heterosis in Pacific oysters [Crassostrea gigas (Thunberg)]. J. Exp. Mar. Biol. Ecol. 233(1): Bayne, B.L., S. Svennson, J.A. Nell. 1999b. The physiological basis for faster growth in the Sydney rock oyster, Saccostrea commercia/is. Biol. Bull. 197: Beiras, R., A. Perez-Camacho, M. Albentosa Influence of food concentration on energy balance and growth perfonnance of Venerupis pullastra seed reared in an open-flow system. Aquaculture 116: Beiras, R., A. Perez-Camacho, M. Albentosa Comparison of the scope for growth with the growth performance of Ostrea edulis seed reared at different food concentrations in an open-flow system. Mar. Biol. 119: Berthelin, C., K. Kellner, M. Mathieu Storage metabolism in the Pacific oyster (Crassostrea gigas) in relation to summer mortalities and reproductive cycle (West Coast of France). Comp. Biochem. Phys. B 125: Boeing, P Use of spray-dried Schizochytrium sp. as a partial algal replacement for juvenile bivalves. Pacific Seafarms International S.A. de C. V., Mazatlan, Sinaloa, Mexico (unpublished). 66

82 Bougrier, S., A.J.S. Hawkins, M. Heral Preingestive selection of different microalgal mixtures in Crassostrea gigas and Mytilus edulis, analysed by flow cytometry. Aquaculture 150: Bradford, M.M A rapid and sensitive method for the quantification of microgram quantities of proteins using the principle of protein-dye binding. Anal. Biochem. 72: Brown, J.R., Multivariate analyses of the role of environmental factors in seasonal and site-related growth variation in the Pacific oyster Crassostrea gigas. Mar. Ecol. Prog. Ser. 45: Brown, M.R., The amino acid and sugar composition of 16 species of microalgae used in mariculture. J. Exp. Mar. Bioi. Ecol. 145: Brown, M.R., G.A. Dunstan, S.W. Jeffrey, J.K. Volkman, S.M. Barrett, J-M. LeRoi The influence of irradiance on the biochemical composition of the prymnesiophyte lsochrysis sp. (Clone T-Iso). J. Phycol. 29:

83 ~- ---~-~ Brown, M.R., S.M. BatTett, J.K. Volkman, S. P. Nearhos, J.A. Nell, G.L. Allan Biochemical composition of new yeasts and bacteria evaluated as food for bivalve aquaculture. Aquaculture 143: Brown, M.R., M.A. McCausland, K. Kowalski The nutritional value of four Australian microalgal strains fed to Pacific oyster Crassostrea gigas spat. Aquaculture 165: Brown, M.R., M.A. McCausland Evaluation of live microalgae and microalgal pastes as supplementary food for juvenile Pacific oysters (Crassostrea gigas). Aquaculture 174(3-4): Brown, M.R., McCausland M.A Increasing the growth of juvenile Pacific oysters Crassostrea gigas by supplementary feeding with microalgal and dried diets. Aquaculture Res. 31 (8-9): Brown, M.R., R. Robert Preparation and assessment ofmicroalgal concentrates as feeds for larval and juvenile Pacific oysters ( Crassostrea gigas). Aquaculture 207:

84 Caers, M., P. Coutteau, P. Sorgeloos Incorporation of different fatty acids, supplied as emulsions or liposomes, in the polar and neutral lipids of the Crassostrea gigas spat. Aquaculture 186: Cigarria, J., Effects of age, size, and season on growth of soft tissue in the oyster Crassostrea gigas (Thunberg, 1793). J. Shellfish Res. 18(1): Cheney, D.P., B.F. MacDonald, R.A. Elston Summer mortality of Pacific oysters, Crassostrea gigas (Thunberg): Initial findings on multiple environmental stressors in Puget Sound, Washington. J. Shellfish Res. 19(1): Coutteau, P., P. Sorgeloos The use of algal substitutes and the requirement for live algae in the hatchery and nursery rearing of bivalve molluscs: An international survey. J. Shellfish Res. 11(2): Coutteau, P., K. Cure, P. Sorgeloos Effect of algal ration on feeding and growth of juvenile manila clam Tapes phillippinarum (Adams and Reeve). J. Shellfish Res. 13(1):

85 Dahlhoff, E.P., J.H. Stillman, B.A. Menge Physiological community ecology: Variation in metabolic activity of ecologically important rocky intertidal invertebrates along environmental gradients. Integr. Comp. Bioi. 42( 4): Enright, C.T., G.F. Newkirk, J.S. Craigie, J.D. Castell Evaluation of phytoplankton as diets for juvenile Ostrea edulis L. J. Exp. Mar. Bioi. Ecol. 96: Elston, R.A., J.H. Beattie, C. Friedman, R. Hedrick, M.L. Kent Pathology and significance of fatal inflammatory bacteraemia in the Pacific oyster, Crassostrea gigas Thuenberg. J. Fish Dis. 10(2): Flores-Vergara, C., B. Cordero-Esquivel, A.N. Ceron-Ortiz, A. Nallely, 8.0. Arredondo Vega Combined effects of temperature and diet on growth and biochemical composition of the Pacific oyster Crassostrea gigas (Thunberg) spat. Aquaculture Res. 35(12): Friedman, C.S., J.H. Beattie, R.A. Elston, R.P. Hedrick. 1991a. Investigation ofthe relationship between the presence of a Gram-positive bacterial infection and summer mortality of the Pacific oyster, Crassostrea gigas Thunberg. Aquaculture 94(1):

86 Friedman, C.S., R.P. Hedrick b. Pacific oyster nocardiosis: Isolation of the bacterium and induction of laboratory infections. J. Inv. Path. 57(1): Garcia, A.S., C.C. Parrish, J.A. Brown Growth and lipid composition of Atlantic cod (Gadus morhua) larvae in response to differently enriched Artemia.franciscana. Fish. Physiol. Biochem. 34: Gouthro, M.A., T.P. Odabasi, E.L. Martin, J.R. Rosowski, J.P. Davis, V.T. Sato Properties of a Schizochytrium (Chromophyta)-based HUF A enrichment and larval feed (Docosa Gold ) suspended in saltwater. J. Shellfish Res. 17( l ): Haigh, R., F.J.R. Taylor, T.F. Sutherland Phytoplankton ecology ofsechelt Inlet, a fjord system on the British Columbia coast. I. General features of the nano- and microplankton. Mar. Ecol. Prog. Ser. 89(2-3): Hanison, P.J, J.D. Fulton, F.J.R. Taylor, T.R. Parsons Review ofthe biological oceanography of the Strait of Georgia: Pelagic environment. Can. J. Fish. Aquatic Sci. 40(7):

87 Harrison, P.J., R.E. Waters, F.J.R. Taylor A broad spectrum artificial seawater medium for coastal and open ocean phytoplankton. J. Phycol. 16: Helm, M.M., I. Laing Preliminary observations on the nutritional value of 'Tahiti lsochrysis' to bivalve larvae. Aquaculture 62: Holland, D.L., P.A. Gabbott A micro-analytical scheme for the determination of protein, carbohydrate, lipid and RNA levels in marine invetiebrate larvae. J. Mar. Biol. Ass. U.K. 51: Holland, D.L., P.J. Hannant Biochemical changes during growth of the spat of the oyster, Ostrea edulis L. J. Mar. Bioi. Ass. U.K. 54: I Holland, D.L., B.E. Spencer Biochemical changes in fed and starved oysters, Ostrea edulis L. during larval development, metamorphosis and early spat growth. J. Mar. Bioi. Ass. U.K. 53: Imai, T., K. Numachi, J. Oizumi, S. Sato Studies on the mass mortality of the oyster in Matsushima Bay II. Search for the cause of mass mortality and the possibility to prevent it by transplantation experiment. Bull. Tohoku Reg. Fish. Res. Lab. 25:

88 J0rgenson, C.B Bivalve filter feeding revisited. Mar. Ecol. Prog. Ser. 142: Knauer, J., P.C. Southgate Nutritional value of a spray-dried freshwater alga, Spongiococcum excentricum, for Pacific oyster (Crassostrea gigas) spat. Aquaculture 146: Knauer, J., P.C. Southgate Growth and fatty acid composition of Pacific oyster (Crassostrea gigas) spat fed microalga and microcapsules containing varying amounts of Eicosapentaenoic acid and Docosahexaenoic acid. J. Shellfish Res. 16(2): Knauer, J., P.C. Southgate A review ofthe nutritional requirements ofbivalves and the development of alternative and artificial diets for bivalve aquaculture. Rev. Fish. Sci. 7(3&4): Knuckey, R.M., M. R. Brown, S.M. Barrett, G.M. Hallegraeff Isolation of new nanoplanktonic diatom strains and their evaluation as diets for juvenile Pacific oysters ( Crassostrea gigas). Aquaculture 211: Laing, I., R-M. Chang Hatchery cultivation of Pacific oyster juveniles using algae produced in outdoor bloom tanks. Aquaculture International 6:

89 Langdon, C.J., M.J. Waldcock The effect of algal and artificial diets on the growth and fatty acid composition of Crassostrea gigas spat. J. Mar. Biol. Ass. U.K. 61: Lassus, P., M. Bardoui1, B. Be1iaeff, P. Masselin, M. Naviner, P. Truquet Effect of a continuous supply of the toxic dinoflagellate Alexandrium minutum Halim on the feeding behaviour of the Pacific oyster ( Crassostrea gigas Thunberg). J. Shellfish Res. 18( 1 ): Lefebvre, S., Barille, L., Clerc, M Pacific oyster (Crassostrea gigas) feeding responses to a fish-farm effluent. Aquaculture 187: Maguire, J.A., P.O. Fleury, G.M. Burnell Some methods for quantifying quality in the scallop Pecten maximus (L.). J. Shellfish Res. 18(1): Manahan, D.T The uptake and metabolism of dissolved amino acids by bivalve larvae. Bioi. Bull. 164: Manahan, D.T., W.B. Jaeckle, S.D. Nourizadeh Ontogenic changes in the rates of amino acid transport from seawater by marine invertebrate larvae (Echinodermata, Echiura, Mollusca). Biol. Bull. 176:

90 Manzi, J.J., N.H. Hadley, M.B. Maddox Seed clam, Mercenaria mercenaria, culture in an experimental-scale upflow nursery system. Aquaculture 54: Matsuyama, Y., H. Usuki, T. Uchida, Y. Kotani Effects ofhannful algae on the early planktonic larvae of the oyster, Crassostrea gigas. Harmful Algal Blooms 2000: McCausland, M.A., M.R. Brown, S.M. Barrett, J.A. Diemar, M.P. Heasman Evaluation of live microalgae and microalgal pastes as supplementary food for juvenile Pacific oysters (Crassostrea gigas). Aquaculture 174: Milke, L.M., J.E. Ward Influence of diet on pre-ingestive particle processing in bivalves II: Residence time in the pallial cavity and handling time on the labial palps. J. Exp. Mar. Bioi. Ecol. 293 : Milke, L.M., V.M. Bricelj, C. C. Panish Comparison of early life history stages of the bay scallop, Argopecten irradians: Effects ofmicroalgal diets on growth and biochemical composition. Aquaculture 260:

91 Milke, L.M., V.M. Bricelj, C.C. Parrish Biochemical characterization and nutritional value of three Pavlova spp. In unialgal and mixed diets with Chaetoceros muelleri for postlarval sea scallops, Placopecten magellanicus. Aquaculture 276: Murata, M., M. Sakaguchi Changes in contents of free amino acids, trimethylamine, and nonprotein nitrogen of oyster during ice storage. Bull. Japan. Soc. Sci. Fish. 52(11): Nell, J.A., J.A. Diemar, M.P. Heasman Food value oflive yeasts and dry yeastbased diets fed to Sydney rock oyster Saccostrea commercia/is spat. Aquaculture 145: O'Connor, W.A., J.A. Nell, J.A. Diemar The evaluation of twelve algal species as food for juvenile Sydney rock oysters Saccostrea commercia/is (Iredale & Roughley). Aquaculture 108: Onal, E., C. Langdon Replacement of living microalgae with spray-dried diets for the marine mussel Mytilus galloprovincialis. Aquaculture 180(3-4):

92 Park, M.S., H.J. Lim, Q. Jo, J.S. Yoo, M. Jeon Assessment of reproductive health in the wild seed oysters, Crassostrea gigas, from two locations in Korea. J. Shellfish Res. 18(2): Parker, R.S., D.P. Selivonchick Uptake and metabolism oflipid vesicles from seawater by juvenile Pacific oysters (Crassostrea gigas). Aquaculture 53: Perdue, J.A The relationship between the gametogenic cycle of the Pacific oyster, C. gigas, and the summer mortality phenomenon in strains of selectively bred oysters. Washington Univ. Sea Grant Program, Seattle, WA. Pemet, F., V.M. Bricelj, C.C. Parrish Effect of varying dietary levels of co6 polyunsaturated fatty acids during the early ontogeny of the sea scallop, Placopecten magellanicus. J. Exp. Mar. Biol. Ecol. 327: Raghukumar, S Bacterivory: a novel dual role for thraustochytrids in the sea. Mar. Bioi. 113: Rice, M.A., G.C. Stephens Uptake and internal distribution of exogenously supplied amino acids in the Pacific oyster, Crassostrea gigas (Thunberg). Aquaculture 66:

93 Robert, R., G. Parisi, L. Rodolfi, B.M. Poli, M.R. Tredici Use of fresh and preserved Tetraselmis suecica for feeding Crassostrea gigas larvae. Aquaculture 192: Robinson, A Dietary supplements for the reproductive conditioning of Crassostrea gigas kumamoto (Thunberg): II. Effects on glycogen, lipid and fatty acid content ofbroodstock oysters and eggs. J. Shellfish Res. 11(2): Rodhouse, P.G., M. O'Kelly Flow requirements ofthe oysters Ostrea edulis L. and Crassostrea gigas Thunb. in an upwelling column system of culture. Aquaculture 22: I Ropert, M., P. Goulletquer Comparative physiological energetics of two suspension feeders: polychaete annelid Lanice conchilega (Pallas 1766) and Pacific cupped oyster Crassostrea gigas (Thunberg 1795). Aquaculture 181: Ross, N.W., K.J. Fitth, A. Wang, J.F. Burka, S.C. Johnson Changes in hydrolytic enzyme activities of naive Atlantic salmon (Salmo salar) skin mucus due to infection with the salmon louse (Lepeophtheirus salmonis) and cortisol implantation. Dis. Aquat. Org. 41:

94 Sakaguchi, M., M. Murata Seasonal variations of free amino acids in oyster whole body and adductor muscle. Nippon Suisan Gakkaishi 55(11): Shumway, S.E., T. L. Cucci, R.C. Newell, C.M. Yentsch Particle selection, ingestion, and absorption in filter-feeding bivalves. J. Exp. Mar. Bioi. Ecol. 91: Somin, J.M., J.M. Deslous-Paoli, 0. Hesse Experimental study of the filtration of clays by the oyster Crassostrea gigas (Thunberg): Adjustment of particle size for best retention. Aquaculture 69(3-4): Soudant, P., M.V. Sanies, C. Quere, J.R. LeCoz, Y. Marty, J. Moal, J.F. Samain, P. Sorgeloos The use of lipid emulsions for sterol supplementation of spat of the Pacific oyster, Crassostrea gigas. Aquaculture 184(3-4): Spencer, B.E., M.J. Akester, I. Mayer Growth and survival of seed oysters in outdoor pumped upwelling systems supplied with fertilized sea water. Aquaculture 55:

95 Spencer, B.E., C.J. Gough The growth and survival of experimental batches of hatchery-reared spat of Ostrea edulis L. and Crassostrea gigas Thunberg, using different mehods of tray cultivation. Aquaculture 13: Tamate, H., K. Numachi, K. Mori, 0. Itikawa, T. Imai Studies on the mass mortality of the oyster in Matsushima Bay. VI. Pathological studies. Bull. Tohoku Reg. Fish. Res. Lab. 25: Thompson, P.A., M. Guo, P.J. Harrison Nutritional value of diets that vary in fatty acid composition for larval Pacific oysters (Crassostrea gigas). Aquaculture 143: Waldcock, M.J., D.L. Holland Fatty acid metabolism in young oysters, Crassostrea gigas: Polyunsaturated fatty acids. Lipids 19(5): Ward, J.E., J.S. Levinton, S.E. Shumway Influence of diet on pre-ingestive particle processing in bivalves I: Transport velocities on the ctenidium. J. Exp. Mar. Biol. Ecol. 293:

96 Webb, K.L., F.L. Chu Phytoplankton as a food source for bivalve larvae. In: G.D. Pruder, C. Langdon and D. Conklin (Editors), Biochemical and physiological approaches to shellfish nutrition. Proc. 2nd Int. Conf. Aquaculture Nutr. World Maricult. Soc. Spec. Publ. No.2, Lousiana State University, LA, pp Whyte, J.N.C Biochemical composition and energy content of six species of phytoplankton used in mariculture of bivalves. Aquaculture 60: Whyte, J.N.C Fatty acid profiles from direct methanolysis oflipids in tissue of cultured species. Aquaculture 75: Whyte, J.N.C., J.R. Englar Seasonal variation in the chemical composition and condition indices of Pacific oyster, Crassostrea gigas, grown in trays or on the sea bed. Can. J. Fish. Aquat. Sci. 39: l 084- l 094. Whyte, J.N.C., K. Sherry, N. Ginther Growth and nutritional condition of Mytilus galloprovincialis fed Schizochytrium. Pacific Biological Station, Fisheries and Oceans Canada, Nanaimo, BC (unpublished). Wilson, J.H The food value of Phaeodactylum tricornutum Bohlin to the larvae of Ostrea edulis L. and Crassostrea gigas Thunberg. Aquaculture 13:

97 Wolfe, G.V The chemical defense ecology of marine unicellular plankton: constraints, mechanisms, and impacts. Bioi. Bull. 198: Wright, D.A., E. W. Hetzel Use of RNA: DNA ratios as an indicator of nutritional stress in the American oyster Crassostrea virginica. Mar. Ecol. Prog. Ser. 25(2):

98 7.0 Appendix 1 83

99 Table 1. Proximate composition of diets used for feeding Crassostrea gigas juveniles for 60 days at 1% dry weight dai 1 Mixed diets were fed at I : 1 (dw:dw) and total oxidation was assumed for calculating total energy. Algamac 3050 and Rotimac are Schizochytrium-based dry feeds (Aquafauna Biomarine, Hawthorne, CA, USA). T-Iso was the fl agellate species lsochrysis galbana (clone T-Iso). Dietary Component T-Iso Algamac 3050 Rotimac Algamac/T-Iso Rotimac/T -lso Protein(%) Carbohydrate (%) Lipid(%) Total Energy (J/mg)

100 Table 2. Proximate composition of diets used for feeding Crassostrea gigas juveniles for 30 days at 1% dry weight dai 1 Mixed diets were fed at 1:1 ( dw:dw) and total oxidation was assumed for calculating total energy. Algamac 2050 and Rotimac are Schizochytrium-based dry feeds (Aquafauna Biomarine, Hawthorne, CA, USA). Thalassiosira nordenskioeldii (Tn.) was the dominant species (>90%) in bloom tank algae used for feeding. Nutritional Component/Diet T. uordeuskioe/tlii AJgamac 2050 Rotimac AM2050/T.u. Rotimac/T.u. Ash content Crude protein(%) Mono/oligosaccharide(%) Polysaccharide (%) Crude lipid(%) Total energy (kj/g)

101 Table 3. The fatty acid composition oftest diets (n = 3; ± S.E.) fed to juvenile Pacific oysters (Crassostrea gigas). The dominant phytoplankton species (>90%) in outdoor grown bloom tank algae cultures was Thalassiosira nordenskioeldii. Rotimac and Algamac 2050 are spray-dried Schizochytrium-based diets (Aquafauna Biomarine, Hawthorne, California, USA). Algamac 2050 Rotimac T-lso Fatty acid Cone. (mg*g 1 OW} %TFA Cone. (mg*g- 1 OW} %TFA Cone. (mg*g- 1 OW} %TFA Saturates 14: ± ± ± iso 15: ± : ± ± : ± ± ± iso 17:0 17: ± : ± ± Total MonoonsatL.Tates 16:1n ± :1n ± :1n6 025 ± :1n5 128 ± :1n9 18:1n ± ± ± :1n ± :1n ± :1n7 Total

102 Table 3 continued. Algamae 3050 Rotimae T-lso Fatty acid Cone. (mg*g- 1 OW) %TFA Cone. (mg*g- 1 OW) %TFA Cone. (mg*g- 1 OW) %TFA Polyunsaturates 16:2n ± ± :2n ± :3n ± ± :4n1 18:2n ± :2n ± ± :3n ± ± :3n ± ± :4n ± :4n ± :2n ± ± ± :3n ± :4n ± :3n ± :5n ± ± :4n ± :4n ± :5n ± ± :5n ± ± :6n ± ± ± Total Unidentified Total

103 Table 4. Individual Pacific oyster (Crassostrea gigas) dry weight after 60 days of feeding. Diets were batch fed for six hours at 1% of the initial dry weight of oysters per tank per day. The reference diet used was Isochrysis galbana (clone T-lso) and Algamac 3050 and Rotimac are spray-dried Schizochytrium-based diets (Aquafauna Biomarine, Hawthorne, California, USA). Diet Algamac 3050 /T- I so ± ± Algamac ± ± Rotimac /T-Iso ± ± Rotimac ± ± T-Iso ± ± Starved (control) ± ±

104 Table 5. Individual Pacific oyster (Crassostrea gigas) AFDW after 60 days of feeding. Diets were batch fed for six hours at I% of the initial dry weight of oysters per tank per day. The reference diet used was!sochrysis galbana (clone T-Jso) and Algamac 3050 and Rotimac are spray-dried Schizochytrium-based diets (Aquafauna Biomarine, Hawthorne, California, USA). Initial AFDW Day60AFDW Diet (%) (%) Chan e (%) Algamac 3050 /T- I so 4.63 ± ± Algamac ± ± Rotimac /T-I so 4.63 ± ± RotimacTM 4.63 ± ± T-Iso 4.63 ± ± Starved (control) 4.63 ± ±

105 Table 6. The fatty acid composition of juvenile Pacific oysters (Crassostrea gigas) fed three test diets (n = 3; ± SE). The dominant phytoplankton species (>90%) in outdoor grown bloom tank algae cultures was Thalassiosira nordenskioeldii (Bloom). Rotimac and Algamac 2050TM (AM 2050 ) are spray-dried Shizochytrium-based diets (Aquafauna Biomarine, Hawthorne, California, USA). DayO Day 30 Rotimac/Bioom Day 30 AM 2050/Bioom Day 30 Bloom Fatty acid I Cone. (mg*g"' OW) %TFAf Cone. (mg*g- OW) %TFAf Cone. (mg*g"' OW) %TFA Cone. (mg*g" ' OW) %TFA Saturates 14: ± ± ± ± iso 15:0 trace ± ± :0 trace ± ± ± : ± ± ± ± iso 17: ± ± ± ± : ± ± ± ± : ± ± ± ± : ± ± ± ± Total Monounsaturates 16:1n9 16:1n ± ± ± ± :1 n ± ± ± ± :1n ± ± ± ± :1 n5 trace 20:1n7 Total

106 Table 6 continued. DayO Day 30 Rotimac/Bioom Day 30 AM 2050/Bioom Day 30 Bloom Fatty acid Cone. (mg*g OW) %TFAI Cone. (mg*g OW) %TFAI Cone. (mg*g OW) %TFAI Cone. (mg*g OW) %TFA Polyunsaturates 16:2n ± ± ± :3n4 trace ± ± ± :3n ± ± :4n ± ± ± :1n9 18:1n7 18:1 n5 18:2n ± ± ± :2n4 18:3n4 18:3n3 18:4n ± ± ± :4n1 20:2n ± ± :5n ± ± ± ± :5n ± ± ± :4n3 22:5n ± ± ± :5n ± ± ± ± :6n ± ± ± ± Total Unidentified Total

107 8.0 Appendix 2 92

108 Figure 1. Photograph of the tank setup for Experiment 1. Each plumbed circuit consisted of a top tank with three upwellers holding juvenile oysters ( Crassostrea gigas) above a reservoir tank using a mechanical pump to recirculate seawater during feeding. Following feeding, ambient 1 11m-filtered seawater was allowed to flow through the top tank. 93

109 100 - Algamac 3050/T-Iso ---v- Algamac m-- Rotimac/T-Iso -<>--- Rotimac 80 --l::r- T -I so -~ -0 -"'C Q) 1/) 60 Q) 0') 1: 1/) Q) 40 0 "'C Q) Q) L Time of feeding (hours) Figure 2. Relative feeding rate of juvenile Pacific oysters ( Crassostrea gigas) as measured by the percentage reduction in cell counts of fi ve diets. Diets were fed at l % the initial dry weight (mg) of oysters per tank for a period of 6 hours. Algamac 3050 and Rotimac are both Schizochytrium-based live algae substitute diets and the reference diet used was Isochrysis galbana (clone T-Iso). 94

110 .---~ v- Algamac 3050 '- Q) Rotimac/T-Iso Algamac 3050/T-Iso -!/) 120 -<>-- Rotimac > l:s- T -I so ~ -0-- Starved "' ) E c: 0) Q) ;: '- Q) -!/) >. 0 ns :::s "'0 :; "'0 c: Number of days fed Figure 3. Mean dry weight (mg ± SE) of individual Pacific oysters (Crassostrea gigas) following 60 days of feeding. Oysters were fed 1% (dw) ofthe initial dry weight of oysters per tank. The reference diet used was Isochrysis galbana (Clone T-Iso) and Algamac 3050 and Rotimac are spray-dried Schizochytrium-based diets (Aquafauna Biomarine, Hawthorne, California, USA). 95

111 5 -.:c 4 C)..._ Algamac 3050/T-Iso ---?---- Algamac Rotimac/T-Iso --o-- Rotimac -----b- T-lso --o-- Starved Q,) 3: ~ "'C 3?!- *-;;;;;::;;;:;;;;:;;;;,=::::sr -... c: Q,)... c: 2 0 (.) c: Q,)... e 1 a Number of days fed Figure 4. Percentage (± SE) protein content of juvenile oysters ( Crassostrea gigas) fed microalgal and live algae substitute diets. Juvenile oysters were batch fed I% of the initial dry weight of oysters per tank for 60 days. The live algae reference diet was Isochrysis gafbana (clone T-Iso) and Algamac 3050 and Rotimac are spray-dried Schizochytrium-based live algae substitute diets (Aquafauna Biomarine, Hawthorne, California, USA). 96

112 6 ----e- Algamac 3050/T -I so -v- Algamac II- Rotimac/T -I so - 5 -o- Rotimac --'= C) Q) ~ ~ 4 "C -C) E - 3 -r::: Q) -r::: 0 (J 2 r::: Q) -0 a.. Q. 1 -l::!r- T -I so -o- Starved ~ ~ ~------~ ~------~ Number of days fed Figure 5. Protein content expressed as mg per individual dry oyster (± SE) following 60 days of feeding test diets. Diets were fed at l % of the initial total oyster dry weight per tank and the reference diet used was lsochrysis galbana (clone T-Iso). Algamac 3050 and Rotimac are spray-dried Schizochytrium-based diets (Aquafauna Biomarine, Hawthorne, California, USA). 97

113 '::' 1.6..c: C) Q) 1.4 ~ "0 ~ 1.2 e;e...., 1.0 -c:: Q) t: "0 0.6 a. - Algamac3050!T-Iso -o- Algamac T- Rotimac!T-Iso - :r- Rotimac -o- T-lso -o- Starved , , ,.-----, Number of days fed Figure 6. Percentage(± SE) lipid content of juvenile oysters (Crassostrea gigas) fed microalgal and live algae substitute diets. Juvenile oysters were batch fed 1% of the initial dry weight of oysters per tank for 60 days. The live algae reference diet was Jsochrysis galbana (clone T-Iso) and Algamac 3050 and Rotimac are spray-dried Schizochytrium-based live algae substitute diets (Aquafauna Biomarine, Hawthorne, California, USA). 98

114 --~ c C) Q) ~ ~ "'C 1.5 -C) E --c: Q) 1.0 -c: 0 (J "'C a....j 0.5. Algamac3050ff-lso -o- Algamac '9'-- Rotimacff-lso ----b- Rotimac -o- T-lso -o- Starved Number of days fed Figure 7. Lipid content expressed as mg per individual dry oyster(± SE) following 60 days of feeding test diets. Juvenile oysters were batch fed 1% of the initial dry weight of oysters per tank for 60 days. The live algae reference diet was Jsochrysis galbana (clone T-Iso) and Algamac 3050 and Rotimac are spray-dried Schizochytrium-based live algae substitute diets (Aquafauna Biomarine, Hawthorne, California, USA). 99

115 - Algamac3050ff-lso -o-- Algamac3050 ~ Rotimacff-lso --t::r-- Rotimac -o- T-lso -o- Starved Number of days fed Figure 8. Percent(± SE) mono/oligosaccharide content of juvenile oysters ( Crassostrea gigas) fed microalgal and live algae substitute diets. Juvenile oysters were batch fed 1% of the initial dry weight of oysters per tank for 60 days. The live algae reference diet was Isochrysis galbana (clone T-Iso) and Algamac 3050 and Rotimac are spray-dried Schizochytrium-based live algae substitute diets (Aquafauna Biomarine, Hawthorne, California, USA). 100

116 Algamac3050/T-Iso -o- Algamac3050 ~ Rotimac/T-Iso -b- Rotimac -o- T-lso -o- Starved 0-0 c: 0 ~ ~------~ ~------~------~------~ Number of days fed Figure 9. Mono/oligosaccharide content expressed as mg per individual dry oyster(± SE) following 60 days of feeding test diets. Juvenile oysters were batch fed 1% of the initial dry weight of oysters per tank for 60 days. The live algae reference diet was Isochrysis gafbana (clone T-Iso) and Algamac 3050 and Rotimac are spray-dried Schizochytrium-based live algae substitute diets (Aquafauna Biomarine, Hawthorne, California, USA). loi

117 Alga mac 3050/T -I so -J: 01 -o- Algamac 3050 Q) Rotimac/T-Iso 3: t::s- Rotimac ~ -D- T-lso -"0 -o- Starved ~ c Q) -c 0 (J 1.5 Q) "0 'i: J: "' (J 1.0 (J "' VI >. 0 a. 0.5 Q) "0 ::::J II.. u Number of days fed Figure 10. Percentage(± SE) polysaccharide content of juvenile oysters ( Crassostrea gigas) fed microalgal and live algae substitute diets. Juvenile oysters were batch fed 1% of the initial dry weight of oysters per tank for 60 days. The live algae reference diet was lsochrysis galbana (clone T-Iso) and Algamac 3050 and Rotimac are spray-dried Schizochytrium-based live algae substitute diets (Aquafauna Biomatine, Hawthorne, California, USA). 102

118 Algamac 3050/T-Iso -..c: en -o-- Algamac 3050 Q),..._ Rotimac/T -I so ~ 2.5 ~ Rotimac ~ -a- T-lso "C -en -a- Starved E s:::: -Q) 1.5 s:::: 0 (J Q) "C 'i: 1.0 ca..c: (J (J ca 0.5 en >. 0 a. 0.0 Q) "C :I '- (.) Number of days fed Figure 11. Polysaccharide content expressed as mg per individual dry oyster(± SE) following 60 days of feeding test diets. Juvenile oysters were batch fed 1% of the initial dry weight of oysters per tank for 60 days. The live algae reference diet was Isochrysis galbana (clone T-lso) and Algamac 3050 and Rotimac are spray-dried Schizochytrium-based live algae substitute diets (Aquafauna Biomarine, Hawthorne, California, USA). 103

119 113 kda 91 kda 53kDa 35 kda 29kDa - 22kDa - Figure 12. Polyacrylamide gel zymography of homogenised tissue from juvenile Pacific oysters (Crassostrea gigas) at Day 0 of a 60 day feeding experiment. Each lane represents tissue from three oysters (n = 3). Low molecular weight marker proteins were used to estimate protease size (kda). 104

120 ~ ---~ Algamac 3050 Roti ma c!t-1 so 113 kda 91 kda Figure 13. Polyacrylamide gel zymography ofhomogenised tissue from juvenile Pacific oysters (Crassostrea gigas) at Day 20 of a 60 day feeding experiment. Low molecular weight marker proteins were used to estimate protease size (kda). Each lane represents tissue from three oysters (n = 3). Algamac 3050 and Rotimac are Schizochytriumbased live algae substitute diets (Aquafauna Biomatine, Hawthorne, California, USA). Isochrysis galbana (clone T-Iso) was used as a control diet ofknown quality for feeding juvenile oysters (C. gigas). Juvenile oysters were batch fed 1% of the initial dry weight of oysters per tank for 60 days and mixed diets were presented as 1:1 (dw:dw) formulations. 105

121 Rotirnac T-lso Starved 91 kda 35 kda 29 kda Figure 14. Polyacrylamide gel zymography ofhomogenised tissue from juvenile Pacific oysters ( Crassostrea gigas) at Day 20 of a 60 day feeding experiment. Low molecular weight marker proteins were used to estimate protease size (kda). Each lane represents tissue from three oysters (n = 3). Rotimac is a Schizochytrium-based live algae substitute diet (Aquafauna Biomarine, Hawthorne, California, USA). lsochrysis galbana (clone T-[so) was used as a reference microalgal diet of known quality for feeding juvenile oysters (C. gigas). Juvenile oysters were batch fed 1% of the initial dry weight of oysters per tank for 60 days and mixed diets were presented as 1:1 ( dw:dw) 106

122 fonnulations. Control oysters were starved to bacterial level (1 1-lffi filtration of incoming seawater). 107

123 Algamac 3050/T-Iso,&\I gam ac Roti mac/t -1 so 1 '13 kda 91 kda 53 kda -- 35kDa -- Figure 15. Polyacrylamide gel zymography ofhomogenised tissue from juvenile Pacific oysters ( Crassostrea gigas) at Day 60 of a 60 day feeding experiment. Low molecular weight marker proteins were used to estimate protease size (kda). Each lane represents tissue from three oysters (n = 3). Algamac 3050 and Rotimac are Schizochytriumbased live algae substitute diets (Aquafauna Biomarine, Hawthorne, California, USA). Isochrysis galbana (clone T-Iso) was used as a control diet of known quality for feeding juvenile oysters (C. gigas). Juvenile oysters were batch fed 1% of the initial dry weight of oysters per tank for 60 days and mixed diets were presented as 1: 1 ( dw:dw) formulations. 108

124 113 kda 91 kda 53 kda 35kDa -- 29kDa., 22kDa ~ J) Figure 16. Polyacrylamide gel zymography ofhomogenised tissue from juvenile Pacific oysters ( Crassostrea gigas) at Day 60 of a 60 day feeding experiment. Low molecular weight marker proteins were used to estimate protease size (kda). Each lane represents tissue from three oysters (n = 3). Algamac 3050 and Rotimac are Schizochytriumbased live algae substitute diets (Aquafauna Biomarine, Hawthorne, California, USA). Isochrysis galbana (clone T-I so) was used as a reference micro algal diet of known quality for feeding juvenile oysters (C. gigas). Juvenile oysters were batch fed 1% of the initial dry weight of oysters per tank for 60 days and mixed diets were presented as 1:1 (dw:dw) fom1ulations. Control oysters were starved to bacterial level (I J.lm filtration of incoming seawater). 109

125 _ Rotimac/T. nordenskioeldii ---v- Algamac2050/T. nordenskioeldii T. nordenskioeldii -C) E J: C) Q,) ~ ~ "C 1.. Q,)... en 25 > ~ :s "C :;: "C 15 c: ~ Number of days fed Figure 17. Dry weight (mg ± SE) per individual Pacific oyster (Crassostrea gigas; n = 30) batch fed once per 24 hours test diets at I% of the initial dry weight of oysters per tank for 30 days. Thalassiosira nordenskioeldii was the dominant phytoplankton species (>90%) in outdoor grown bloom tank cultures. Rotimac and Algamac 2050 are spray-dried Schizochytrium-based live algae substitute diets (Aquafauna Biomarine, Hawthorne, California, USA). 110

126 J: 12 C) Q) ~ 10 ~ - "C ~ 0 8 -c:: Q) -c:: 0 6 (.) (.) c:: 4 nj C) ~ 0 2._ Rotimac/T. nordenskioeldii ---y-- Algamac2050/T. nordenskioeldii T. nordenskioe/dii Number of days fed Figure 18. Percentage organic content or ash-free dry weight(± SE) of juvenile Pacific oysters (Crassostrea gigas) batch fed test diets at 1% of the initial dry weight of oysters per tank for 30 days. Thalassiosira nordenskioeldii was the dominant phytoplankton species (>90%) in outdoor grown bloom tank cultures. Rotimac and Algamac 2050 are spray-dried Schizochytrium-based live algae substitute diets (Aquafauna Biomarine, Hawthorne, California, USA). I I I

127 8 -.s:: C) Q) 7 ~ ~ "C ~ - 6 -s::: Q) -s::: 0 0 s::: 5 Q) -0 ~ a. 4._ Rotimac/T. nordenskioeldii -r:::r- Algamac 2050/T. nordenskioeldii - :1- T. nordenskioeldii 3 ~ ,, Number of days fed Figure 19. Percentage(± SE) crude protein content per individual dry Pacific oyster (Crassostrea gigas) batch fed once per 24 hours test diets at 1% of the initial dry weight of oysters per tank for 30 days. Thalassiosira nordenskioeldii was the dominant phytoplankton species (>90%) in outdoor grown bloom tank cultures. Rotimac and Algamac 2050 are spray-dried Schizochytrium-based live algae substitute diets (Aquafauna Biomarine, Hawthorne, California, USA). 112

128 -- -, 2.0 Q) ~ ~ "'0 0, 1.5 E e- Rotimac/T. nordenskioeldii --v- Algamac 2050/T. nordenskioeldii ---{]--- T. nordenskioeldii ~ ~------~------~ ~------~ Number of days fed Figure 20. Crude protein content (mg ± SE) per individual dry Pacific oyster (Crassostrea gigas) batch fed once per 24 hours test diets at l% of the initial dry weight of oysters per tank for 30 days. Thafassiosira nordenskioefdii was the dominant phytoplankton species (>90%) in outdoor grown bloom tank cultures. Rotimac and Algamac 2050 are spray-dried Schizochytrium-based live algae substitute diets (Aquafauna Biomarine, Hawthorne, California, USA). 113

129 z !: Ol Q) ~ ~ "C ~ -c:.! 0.15 c: 0 (.) Q) "C 'i: ~ !: (.) (.) ~ en 0 Ol Rotimac/T. nordenskioeldii s;;j- Algamac 2050/T. nordenskioeldii T. nordenskioeldii 0-0 c: 0 ::.:! , , , , Number of days fed Figure 21. Percentage(± SE) crude mono/oligosaccharide content per individual dry Pacific oyster (Crassostrea gigas) batch fed once per 24 hours test diets at I% of the initial dry weight of oysters per tank for 30 days. Thalassiosira nordenskioeldii was the dominant phytoplankton species (>90%) in outdoor grown bloom tank cultures. Rotimac and Algamac 2050 are spray-dried Schizochytrium-based live algae substitute diets (Aquafauna Biomarine, Hawthorne, California, USA). 114

130 ~ 0.07 C) Q) 3: 0.06 ~ "C -C) E o.o5 -c Q) c Q) "C 0.03 'i: ns J: 0 0 ns /) 0 C) 0-0 c _ Rotimac/T. nordenskioeldii ---v- Algamac 2050/T. nordenskioeldii ----1:1- T. nordenskioeldii ::!: t , , Number of days fed Figure 22. Crude mono/oligosaccharide content (mg ± SE) per individual Pacific oyster (Crassostrea gigas) batch fed once per 24 hours test diets at 1% of the initial dry weight of oysters per tank for 30 days. Thalassiosira nordenskioeldii was the dominant phytoplankton species (>90%) in outdoor grown bloom tank cultures. Rotimac and Algamac 2050 are spray-dried Schizochytrium-based live algae substitute diets (Aquafauna Biomarine, Hawthorne, California, USA). 115

131 3.5 -"- a> -(/) > ~ "C ";! r:: 0 :;: E 2.0 -r:: a> (,) g 1.5 (,) a> "C 'i: 1.0 ra.r:: (,) (,) ~ 0.5 ~ 0 a Rotimac/T. nordenskioeldii --'\7- Algamac 2050/T. nordenskioeldii -D- T. nordenskioeldii , ,, , Number of days fed Figure 23. Percentage(± SE) crude polysaccharide content per individual Pacific oyster (Crassostrea gigas) batch fed once per 24 hours test diets at 1% of the initial dry weight of oysters per tank for 30 days. Thalassiosira nordenskioeldii was the dominant phytoplankton species (>90%) in outdoor grown bloom tank cultures. Rotimac and Algamac 2050 are spray-dried Schizochytrium-based live algae substitute diets (Aquafauna Biomarine, Hawthorne, California, USA). 116

132 1.0 _._ Rotimac/T. nordenskioeldii -v- Algamac 2050/T. nordenskioeldii ~ T. nordenskioe/dii , Number of days fed Figure 24. Crude polysaccharide content (mg ± SE) per individual Pacific oyster (Crassostrea gigas) batch fed once per 24 hours test diets at 1% of the initial dry weight of oysters per tank for 30 days. Thalassiosira nordenskioeldii was the dominant phytoplankton species (>90%) in outdoor grown bloom tank cultures. Rotimac and Algamac 2050 are spray-dried Schizochytrium-based live algae substitute diets (Aquafauna Biomarine, Hawthome, California, USA). 117

133 Rotimac/T. nordenskioeldii...sv- Algamac 2050/T. nordenskioeldii ----r.- T. nordenskioeldii --.c: C) 1.2 Cl) ~ ~ 1.0 "C ~ c: Cl) -c: (.) "C Q Cl) "C :::::s '- (.) Number of days fed Figure 25. Percentage(± SE) crude lipid content per individual Pacific oyster ( Crassostrea gigas) batch fed once per 24 hours test diets at 1% of the initial dry weight of oysters per tank for 30 days. Thalassiosira nordenskioeldii was the dominant phytoplankton species (>90%) in outdoor grown bloom tank cultures. Rotimac and Algamac 2050 are spray-dried Schizochytrium-based live algae substitute diets (Aquafauna Biomarine, Hawthorne, California, USA). 118

134 -{I- -.s:::: C) Q) ~ ~ "C -C) E Rotimac/T. nordenskioeldii ~ Algamac 2050/T. nordenskioeldii T. nordenskioeldii c: 0.6 Q) -c: 0 (.) "C 0.4 a. Q) "C ::J 0.2 ~ u r , Number of days fed Figure 26. Crude lipid content (mg ± SE) per individual Pacific oyster (Crassostrea gigas) batch fed once per 24 hours test diets at 1% of the initial dry weight of oysters per tank for 30 days. Thalassiosira nordenskioeldii was the dominant phytoplankton species (>90%) in outdoor grown bloom tank cultures. Rotimac and Algamac 2050 are spray-dried Schizochytrium-based live algae substitute diets (Aquafauna Biomarine, Hawthorne, California, USA). 119

135 ~~~~ Rotlmac/T. nordenskioeldll Algamac 2050/T. nordenskloeldli 70 i 00 ~ i!10 :3. '"' 0 Lipid (J/mg) o Mono (Jim g) : :~!e~~~~~g ) 70 0 Lipid (J/mg) o Mono (J/mg) Poly (J/mg) Protein (J/mg), " " Number of days. fed 30,,, Number of days fad "' 0 Lipid (J/mg) 0 Mono (J/mg) Poly (Jim g) Protein (J/mg) T. nordenskioeldii, " " "' Number of days fed Figure 27. Energy equivalent of the proximate composition of the organic fraction of juvenile Crassostrea gigas fed test diets for 30 days. Total components is the sum of tested components (protein, lipid, polysaccharide and mono/oligosaccharides; Joules mg- 1 oyste{ 1 ). Oysters were batch fed once per 24 hours test diets at 1% of the initial dry weight of oysters per tank for 30 days. Thalassiosira nordenskioeldii was the dominant phytoplankton species (>90%) in outdoor grown bloom tank cultures. Rotimac TM and Algamac 2050 are spray-dried Schizochytrium-based live algae substitute diets (Aquafauna Biomarine, Hawthorne, California, USA). 120

136

137

138

139